169results about How to "General" patented technology

The invention discloses a method for linking accounts in a plurality of OSNs (On-line Social Networks) on the basis of node similarity. The friend relationship is extracted from OSN website accounts for linking the plurality of accounts, belonging to one real user, in different OSN platforms. The method comprises five parts: a data preprocessing part performs preprocessing on an account node relationship diagram of the OSNs; a node sequence extraction part obtains an account node sequence set through random walk; an account vector representation part generates a vector model of each account through a word to vector tool word2vec; a linear conversion matrix calculation part obtains a linear conversion matrix W from one OSN to another OSN by a gradient descent method; a linking account obtaining part maps the account in one OSN to a coordinate space of another OSN; and through similarity measure and threshold value screening, the linking accounts or a candidate set corresponding to all of the accounts are obtained. Errors of the account linking result caused by unreal feature information of the accounts are avoided, so that the account linking robustness is improved.

The invention discloses a motion capture method based on inertia and optical measurement fusion, and relates to the field of a sensing technology and computer vision, in particular to capture and analysis of motion processes of one or more objects. An inertia measurement technology and an optical capture technology are combined. The inertia measurement technology has the characteristics that the inertia measurement technology is insusceptible to the external environment, but problems of error accumulation, data drifting, poorer dynamic tracking performance and the like exist during measurement; the optical motion capture technology is high in accuracy but is limited by the environment such as illumination, colors and the like easily and interfered by shielding. The motion capture system and method integrate advantages of the inertia measurement technology and the optical capture technology, are low in cost, high in accuracy, small in error and insusceptible to the environment and are not interfered by shielding.

The invention discloses a millimeter wave folding-type reflective array antenna integrated with a plane feed source. The millimeter wave folding-type reflective array antenna is of a layered structure and is sequentially provided with a polarization grid, a reflective array and the plane feed source from top to bottom. On the basis of a plane substrate integration waveguide structure, a substrate integration waveguide gap array antenna is used as the feed source for conducting feed on folding-type reflection, phase compensation is obtained on a reflection face, spherical waves are converted to plane waves, and therefore the antenna which is high in gain and efficiency and stable in beam pointing is realized. Under the Q-LINKPAN application background and directing at the development demands of a plane-integration and miniaturized long-distance wireless communication system, the low-section antenna which is high in gain and efficiency and capable of being integrated with a plane millimeter wave circuit is realized, the millimeter wave folding-type reflective array antenna has the advantages of being simple in structure, compact in size and low in cost, and the requirements for plane circuit integration are met.

The invention discloses a mechanical equipment fault intelligent diagnosis method based on a generation model under small sample data. The mechanical equipment fault intelligent diagnosis method comprises: carrying out zero-mean-value standardization preprocessing on a small number of obtained mechanical signals; Establishing a composite network for mechanical signal generation; Training a generative adversarial network model in an adversarial manner in combination with a Wasserstein distance and a gradient punishment method; Establishing a deep convolutional neural network model for classifying the operation states of the mechanical equipment by using the mechanical signals; And by combining the generative adversarial composite neural network model and the deep convolutional neural network model, training two networks by using a small amount of real mechanical signals, and finally realizing intelligent fault diagnosis of the mechanical equipment under small sample data. The method hasthe advantages of being good in mechanical signal feature extraction effect, high in state classification accuracy and good in mechanical signal data expansion performance.

The invention relates to a cache method for a file system with changeable data block length, belonging to the technical field of computer data storage. The method of the invention comprises the following steps: intercepting operating commands from a computer application program by a driving module of a user state file system, wherein the operating commands comprises opening the existing files, reading data from the opened files, writing data into the opened files, modifying the file attributes and closing the opened files; operating a cache management module in a daemon process mode; and temporarily saving partial data which is read and written recently in each currently opened file into data blocks with random length in the internal storage of a computer, thereby reducing the frequency of access of the file system to external storage devices, and improving the efficiency and throughput of accessing data in the file system. The method of the invention can enable the data reconstruction and data segmenting processes to be transparent to the user and computer application programs, thereby enabling the file system to support a repeating data deleting function under the condition that the compatibility is not affected, and effectively improving the utilization efficiency of the physical storage space.

The invention discloses an ancient Chinese automatic translation method based on multi-feature fusion. The method comprises the following steps: 1) collecting a text, modern text translation data of the text, a text word list and modern Chinese monolingual corpus data; And 2) cleaning the data and constructing an ancient Chinese parallel corpus by using a sentence alignment method. And 3) carryingout word segmentation on the modern text and the ancient text by using a Chinese word segmentation tool; 4) performing topic modeling on the ancient text corpus to generate topics-Word distribution and word-Subject conditional probability distribution 5) using the modern Chinese monolingual corpus to train to obtain a modern Chinese language model; And obtaining an aligned dictionary by using ancient Chinese parallel corpora. 6) on the basis of the attention-based recurrent neural network translation model, fusing statistical machine translation characteristics such as a language model and analignment dictionary, and using an ancient Chinese parallel sentence pair and a word topic sequence training model, and 7) inputting a to-be-translated text by a user, and obtaining a modern text translation by using the model obtained by training in the step 6).

The invention discloses a graphene based terahertz broadband adjustable wave absorption device, and relates to a wave absorption device. The graphene based terahertz broadband adjustable wave absorption device is of a three-layer structure which comprises a periodic mesh-shaped graphene layer, a dielectric layer and a metal layer from top to bottom, and is characterized in that the horizontal period of a unit structure of the periodic mesh-shaped graphene layer is Px, the longitudinal period of the unit structure is Py, the graphene width has a characteristic of continuous gradual change modulation, the edge of the periodic mesh-shaped graphene layer has a sine curve, a cosine curve, a parabolic curve, an elliptic curve or other curves capable of realizing graphene width gradual change modulation, the maximum width Wmax is consistent with the horizontal period Px, and the minimum width Wmin ranges between 0 and the horizontal period Px; the dielectric layer is made of an insulator such as silicon dioxide, and the thickness td of the dielectric layer is associated with the wave absorption frequency band; and the metal layer is made of a good conductor such as gold or silver, and the thickness tm of the metal layer is greater than the skin depth of incident waves. The graphene based terahertz broadband adjustable wave absorption device can realize broadband strong absorption for terahertz waves, the absorption rate of incident broadband terahertz waves is adjusted on a large scale, and the structure is simple.

The present invention discloses a DNA molecular chain displacement reaction based method for extracting CRNs for realizing combinational logic. The method comprises the following steps of: step 1, according to a set logic function, obtaining a truth table that reflects input and output logic relationships, and obtaining, through mapping, a proforma chemical reaction network from the truth table; and step 2, simplifying the proforma chemical reaction network obtained in the step 1 by using a Karnaugh map simplification approach, to obtain a proforma chemical reaction network for realizing a combinational logic function. The method provided in the present invention is mainly used for obtaining the chemical reaction network that represents logical relationships in a DNA molecular chain displacement reaction; the CRNs is obtained by mainly using the truth table in electronics and the Karnaugh map simplification approach; CRNs that indicate logical relationships can be obtained without considering the specific circuit architecture; the CRNs are simplified; and energy consumption is effectively reduced. Meanwhile, the method provided in the present invention is more general, straightforward and more convenient in operation.

The invention discloses an RBF neural network adaptive control method for multiple single mechanical arms. The method comprises the steps of: approximating an unknown nonlinear function in a mechanical arm system by a neural network; introducing a dynamic surface technique to design a first-order filter to solve the computational explosion caused by repeated derivation on a control in a backstepping method; and processing the unknown parameters and output limitation by a Nussbaum function and an obstacle Lyapunov function. The method does not require an accurate mechanical kinetic model, can completely eliminate the output error caused by unknown kinetic parameters and random interference, solves the problem that a model-based multi-single mechanical arm control scheme depends on a precise kinetic model, and improves the dynamic performance of the mechanical arm and the trajectory tracking accuracy of a joint space. Finally, the feasibility and effectiveness of the control method are verified by simulation examples.

The invention discloses a pre-coding and power allocation joint optimization method in a millimeter wave NOMA multi-beam system down link. The pre-coding and power allocation joint optimization methodcomprises the following steps: (1) a base station obtains downlink channel information of a user; (2) performing grouping pairing on the user according to the obtained downlink channel information ofthe user; (3) designing a set of codebook, and transforming the pre-coding and power allocation joint optimization problem into a convex optimization problem; (4) solving the transformed convex optimization problem through a iteration method and obtaining a selection factor vector and a power allocation factor vector; and (5) constructing a new codeword as a pre-coding matrix using the selectionfactor and the codebook. The invention solves the joint optimization problem of millimeter wave NOMA pre-coding and power allocation in a scenario in which the base station transmits multiple beams, which is more general.

The invention relates to a node current injection method-based modeling method of a unified power flow controller. For the unified power flow controller, a power flow control model of the unified power flow controller is established by the following steps: step one, establishing an equivalent circuit of the unified power flow controller; step two, for the equivalent circuit obtained by the step one, establishing an equivalent model expressed by a node current injection method so as to obtain an injected current of a parallel side node s and an injected current of a serial side node m of the unified power flow controller; and using the equivalent model and a corresponding injected current computational formula to obtain the power flow control model of the unified power flow controller. The method has the high adaptability; and the bus voltage control effect is good.

The invention belongs to the technical field of image retrieval, and relates to a self-adaptation Hash rearrangement method for image retrieval, in particular to an image Hash method for carrying out content-based image retrieval. The method adopts the mapping and then sequencing Hash rearrangement method, and comprises the following steps: high-dimensional visual feature vectors of images in a training library are abstracted, a proper Hash method is selected to map the high-dimensional visual feature vectors into Hash codes, and specific class weight vectors are generated for each class of images; the hamming distance between the Hash codes of the retrieved images and the Hash codes in the training library is calculated, and retrieval results are returned in an ascending order; and according to the retrieval results, the self-adaptation weight vectors of the retrieved images are calculated, the hamming distance is weighted by utilizing the structures of the self-adaptation weight vectors of the retrieved images, the returned images are rearranged by utilizing the weighted hamming distance, and more accurate retrieval results are obtained. The self-adaptation Hash rearrangement method calculates specific weights according to the retrieved images, has universality, and remarkably improves the retrieval effect without increase of calculation complexity.

The present invention discloses a method for concluding and classifying online comments of a hospital, and aims to conclude various online evaluations of the hospital by patients, summarize the overall impression of the hospital, and implementing classification of the comments. The method comprises the following steps of first dividing crawled Chinese comment content into a plurality of short sentences by using regular expression; then performing word segmentation and part-of-speech tagging of each short sentence, and merging adjacent nouns into a noun; replacing adjectives on basis of synonym dictionaries with the principle that low-frequency words are replaced by high-frequency words; storing the obtained noun+adjective phrases into a database, and simultaneously establishing mapping relation among the phrases, the short sentences and the comments; and inputting the tile of the hospital, and selecting a plurality of high-frequency phrases from the database as the overall evaluation of the hospital. According to the method, the online comments of the hospital can be simplified, the trouble of reading various comments is saved, and the comments can be read by classification conveniently, thereby facilitating healthcare service choice and hospital choice.

A permission control method for calling of various service type based on adaptive control, which pertains to wireless network technology. The invention provides differentiating method of various service type based on QoS requirements of different service type in wireless network system; establishes model of queue with priority, proposes resource combating method of different service type when system resource is scarce, and provides general solving procedure of call dropping probability when service switching and blocking probability of new call by analysis of Markov birth and death process; establishes adaptive fuzzy degrade control model, proposes degrade control method of the current call when system resource is scarce, realizes dynamic balance of user QoS and system resource utility; finally, establishes optimal allocation model of degrade amplitude objects by non-linear programming algorithm. Three models proposed by the invention constitutes a close-loop control system together, which is excellent in completeness and robustness, and easy to implement.

A graphene multi-resonant structure-based polarization wideband-independent terahertz wave absorber relates to a terahertz wave absorber. The graphene multi-resonant structure-based polarization wideband-independent terahertz wave absorber comprises a five-layer structure, wherein the five-layer structure is sequentially a periodic graphene multi-resonant mesh-shaped structural layer, an upper dielectric layer, a bias layer, a lower dielectric layer and a metal layer from top to bottom, horizontal period and a longitudinal period of a unit structure of the periodic graphene multi-resonant mesh-shaped structural layer both are in square shapes, the graphene multi-resonant structure comprises square graphene, a one fourth round disc and a rectangular strip, the square graphene is formed by connecting middle points of four edges of a periodic square shape, the one fourth round disc has radius of four corners of the periodic square shape, and the rectangular strip has width on four edges of the periodic square shape. The graphene multi-resonant structure-based polarization wideband-independent terahertz wave absorber is simple in structure, universality and can be used for absorbing anelectromagnetic wave of an intermediate infrared band, a far infrared band, visible light or other bands by scale conversion.

The invention relates to an LED (light emitting diode) driving power supply with a high power factor and without electrolytic capacity. The LED driving power supply mainly comprises an alternating current input power supply, a monophase uncontrolled rectifier bridge, a main DC (direct current) / DC conversion circuit, an auxiliary DC / DC conversion circuit, an energy storage capacitor and a filter capacitor. The LED driving power supply is mainly and technically characterized in that the auxiliary DC / DC conversion circuit is merged into the input end of the main DC / DC conversion circuit, and has the functions of balancing an instantaneous power between input and output to be without the electrolytic capacity and compensating a current wave shape on the input end to realize the high power factor. By the adoption of the driving power supply, the defect that a three-time harmonic wave injection method is limited by the power factor is made up, and a method of double frequency current wave shape fitting can be further adopted, thus the acquisition of a reference current wave shape is further simplified, and the controlling is facilitated. As a voltage ripple on the energy storage capacitor has no requirement, needed capacity values of the energy storage capacitor and the filter capacitor are extremely small, a film capacitor or a ceramic capacitor and the like can be directly adopted, the driving power supply has the advantages of long service life and the like, and is suitable for places of a public driving power supply of an LED with large power.

The invention discloses a cross-media Chinese herbal medicine plant image searching method based on deep learning. The method comprises the following steps that 1, on the basis of OCR, text structural processing is performed, and plant classification description characters are extracted from books such as 'plant taxonomy'; 2, a Chinese word segmentation tool is used for performing preprocessing including word segmentation and stop word removal on all plant classification description characters; 3, a word2vec algorithm is used for generating word vectors according to a description text; 4, Fisher Vector is used for encoding the description text; 5, a convolutional neural network is used for performing training on an image set, and the network structure is made to converge to the best state; 6, the full last but one connection layer output of the convolutional neural network is extracted as an image feature vector; 7, textual features and image features are fused; 8, a linear kernel SVM classifier is used for performing model training; 9, during user retrieval, images can be input, plant texts can be described, and then a final image retrieval result is obtained through the steps 4, 6, 7 and 8.

The invention provides a bus-type lithium battery pack balance system prediction control method, which comprises steps: 1, a state space model for the lithium battery pack balance system is built; 2, an outer loop controller adopts the prediction control method, the balance speed and the balance efficiency are considered at the same time, and the power of a single battery in the battery pack is predicted and controlled; and 3, an inner loop controller adopts a PI controller form to realize that power of the battery changes according to a scheduled track through controlling the balance current of the single battery and the energy bus voltage.

The invention discloses a self-learning-based variable working condition mechanical fault intelligent diagnosis method under a small sample. The method comprises the steps of performing standardized preprocessing on a mechanical signal sample of mechanical equipment in a variable working condition operation state; constructing and training a generative adversarial network, and enhancing a trainingdata set by generating a mechanical signal sample; constructing a self-learning network model; pre-training a self-learning network model by using the mechanical signal samples generated by the generative adversarial network and not more than 30 real mechanical signal samples; constructing a mechanical equipment state classification and recognition network model; and the network weight of the mechanical equipment state classification and recognition network model is finely adjusted by using not more than 30 real mechanical signal samples and the corresponding state labels, and the mechanicalequipment state classification and recognition network model after fine adjustment of the weight can realize intelligent diagnosis of mechanical faults under variable working conditions. The method has the characteristics of high diagnosis result accuracy and strong anti-interference capability.

The invention discloses an organic Rankine cycle system fault detection method based on a generalized cross-entropy-DPCA (Dynamic Principal Component Analysis) algorithm, and relates to the field of non-Gaussian random system fault diagnosis. The generalized cross-entropy-DPCA algorithm is mainly utilized; after normalization processing is carried out, a generalized Gaussian kernel function is adopted to replace a Gaussian kernel function to define the cross-entropy again to obtain a generalized cross-entropy-DPCA performance index, and the performance index is optimized to solve an optimal direction matrix; meanwhile, a confidence coefficient [Alpha]is set to solve control limits for a probability density function integral; after the direction matrix is solved, fault data under a fault working condition is adopted and is subjected to the normalization processing; then, the fault data is calculated by the direction matrix to obtain the SPE statistical magnitude and the T2 statistical magnitude of the fault data model; and the SPE statistical magnitude and the T2 statistical magnitude are compared with the calculated control limits to detect whether the system fails or not. By use of the organic Rankine cycle system fault detection method, a generalized Gaussian kernel is adopted, the method exhibits better universality, and in addition, fault detection accuracy is improved.

The invention discloses a generation type dialogue abstract method integrated with common knowledge, and belongs to the field of natural language processing. According to the invention, the problems of inaccurate generated dialogue abstract and low abstraction caused by the fact that the conventional generative dialogue abstract method does not utilize common knowledge are solved. The method comprises the following steps: acquiring a common knowledge base ConceptNet and a dialogue abstract data set SAMSum; introducing tuple knowledge into a dialogue abstract data set SAMSum by utilizing the obtained common knowledge base ConceptNet, and constructing a heterogeneous dialogue diagram; training the dialogue heterogeneous neural network model constructed in the step 3, and generating a final dialogue abstract from a section of dialogue through the trained dialogue heterogeneous neural network model. The method is applied to generation of the dialogue abstract.

The invention provides a method for acquiring an ADL (architecture description language) and a modeling tool. The method comprises the following steps: firstly, formulating a DPOSPL (data process oriented software product line) ADL metamodel for describing the product line variability; then introducing variable definitions and an assignment syntax in the DPOSPL ADL; and finally, formulating a description specification on the DPOSPL ADL in accordance with the restriction of an architectural model, wherein an architecture description language supports the transformation from a demand characteristic model subjected to domain analysis to an architecture model, thus quickly developing an architecture of actual products aiming at specific demands directing at variable points in the demand characteristic model.

The invention discloses a method for identifying the brake performance of a safety monitoring system for motor vehicles. The method comprises the following steps: obtaining operating variables of acceleration, pressure and temperature input and brake performance parameter output; analyzing a redundant information relation between the operating variables by using a principal component analysis method so as to eliminate redundant data; and determining the structure of a recurrent wavelet ELman neural network, and obtaining a plurality of accurate values of brake performance parameters for the operating process by using an identification method. In the invention, through carrying out identification processing on obtained brake performance sensing quantities, accurate sensing information can be provided for a rear-end safety status forecasting module; the method can be used for carrying out identification on operating multi-parameter input information of motor vehicles, therefore, the method has strong learning and generalization capabilities; and by using the method disclosed by the invention, inevitable error accumulation caused by a specific computation model can be shielded, therefore, the accuracy of safe-operating brake performance monitoring parameters of motor vehicles can be effectively improved.

The invention discloses a method for analyzing compatibility of a satellite navigation system based on a code tracking frequency spectral separation coefficient, relates to a method for analyzing compatibility of a satellite navigation system, and belongs to the field of satellite navigation. The method comprises the following steps: using a TOA module to analyze influence of a code phase evaluated error on results of a TOA estimator and a filter, to obtain relations of a variance of a smooth TOA estimator and an unsmooth TOA estimator; analyzing influence of the code phase evaluated error and a carrier frequency evaluated error on a code tracking error, to obtain a code tracking error of a code tracking loop advanced-lag process under interference; determining a spectral separation coefficient based on the code tracking; determining equivalent carrier-to-noise ratio of the satellite navigation system in code tracking; and using the obtained equivalent carrier-to-noise ratio to assess influence of navigation interference signals on a target signal in a signal code tracking state of the satellite navigation system, determining the compatibility of the satellite navigation system. The method can improve assessment precision of compatibility among satellite navigation systems, and improve anti-interference among signals of the satellite navigation systems.

The invention discloses a Chinese herbal medicine plant picture retrieval method based on a multi-feature fusion BOW (Bag-of-Words) model. The method comprises the following steps of firstly, performing normalization processing on the dimension of Chinese herbal medicine pictures; then, extracting SIFT (Scale-invariant Feature Transform) features and Opponent SIFT features based on dense sampling on the Chinese herbal medicine pictures subjected to normalization processing; then, clustering the features extracted in the former step by a k-means clustering algorithm to obtain a visual dictionary, wherein the clustering center is a visual word; next, expressing the visual word in the visual dictionary for local features by local linear code to obtain a corresponding coding vector; then, performing pooling processing on the code vector corresponding to local features of the Chinese herbal medicine pictures by using a maximum value pooling method to obtain picture features; and finally, inputting the picture features into a trained SVM (Support Vector Machine) classifier to obtain corresponding categories; and returning the pictures in the corresponding categories to a user.

The invention belongs to the field of multi-agent system control and particularly relates to a safety consistency control method for a multi-agent system. For any normal agent in the multi-agent system, the control method comprises steps: a sequence containing own information and neighbor delay information is ascended, the state values in the sequence that are not in the safety domain are removed, and a new sequence after preprocessing is thus obtained; the mean and the standard deviation of the new sequence are calculated with one standard deviation away from the mean as a constraint condition; the state values in the new sequence that are not in the constraint condition are eliminated until all kept state values in the new sequence meet the constraint condition; and the kept neighborinformation is used for own state updating. Malicious value judgment can be carried out on al neighbor information of a normal agent, and the anti-attack capability of the system is enhanced.

Disclosed is a motion parameter estimation system for orbital maneuver targets based on a transient relative model, which belongs to the mechanics and control fields, wherein: an input information preprocessing module preprocesses the input information for providing the compatible real-time measurement and time-varying parameter information to a transient relative model module, the transient relative model module updates the input measurement content and time-varying parameters of the transient model in real time at the moment of measurement and sampling for feeding the updated model into a relative state estimation module, and the relative state estimation module obtains a relative state estimated value, one path of the relative state estimated value serves as the relative state output and the other path thereof is fed into a target maneuver identification module to make a one-step prediction of relative state through sampling delay, and the target maneuver estimated value output is obtained by using the maneuver parameter variable equation, thereby judging relative motion conditions accurately in real time. The system is also applicable to targets running on unknown tracks in unknown maneuver conditions, thereby benefiting achievements of such space tasks as long-term autonomous monitoring and tracking for maneuvering targets.

The invention belonging to the energy optimization field of the energy storage system relates to a distributed-extreme-value-search-algorithm-based energy storage system and an energy optimization management method thereof. The method comprises: operating state information of a current energy storage unit and an adjacent energy storage unit are detected; on the basis of the operating state information, whether power generated by a distributed power generating unit meets the need of a load is determined; if so, whether an energy storage system needs to be charged or a distributed power supply provides power for a large grid is determined intelligently based on an economic dispatch requirement; if not, whether the energy storage system needs to be discharged or the load purchases power fromthe large grid is determined intelligently based on an economic dispatch requirement; if the charging or discharging of the energy storage units is selected, a distributed extreme value search algorithm is used for control an execution unit; whether the charge quantities of the energy storage units meet an expected standard is determined; if so, the operation is completed; and if not, the controller is used continuously to carry out optimized controlling. Therefore, the input and output of the energy storage system are controlled in real time; stability of the system is ensured; and the energylosses are reduced effectively.

The invention discloses a voltage sag level comprehensive evaluation method comprises based on an IEEE Std 1564-2014 standard, selecting amplitude, frequency, energy and severity indexes to constructa node index vector; trending and nondimensionalizing the original index data by using a reverse inverse transformation method and an averaging method so as to preprocess the indexes; constructing a determining matrix based on an analytic hierarchy process, and subjectively weighting the indexes; objectively evaluating the indexes by using a principal component analysis method based on a TOPSIS model; and comprehensively evaluating the node sag level by using combination weighting. The method solves the excessive information redundancy level between sag indexes, combines the advantages of subjective experience and objective information by the combination weighting method, and overcomes the defect that a traditional voltage sag evaluation method just uses a single weighting method and cannot fully describe the defects of the sag feature. A provided test method can describe the performance in all aspects of the comprehensive evaluation method accurately and comprehensively.

The invention provides an electrochemical energy storage power station planning method considering the cycle life of an energy storage battery, and belongs to the technical field of power system planning and operation. The method comprises the steps of firstly, establishing and solving an electrochemical energy storage power station stochastic programming model under a typical scene set, and obtaining an initial programming result of a to-be-programmed power station; secondly, performing annual daily operation simulation on the power system by utilizing the current planning result, constructing an energy time sequence curve of the electrochemical energy storage power station, and calculating the actual cycle life of the electrochemical energy storage power station, thereby obtaining the actual service life of the electrochemical energy storage power station; and adjusting the current planning result by calculating the parameter sensitivity of the power system to the electrochemical energy storage power station until a final planning result conforming to the error tolerance is output. According to the method, the accurate cycle life model of the energy storage battery and various uncertain energy storage application scenes are considered, the accuracy of the planning scheme is improved, and the obtained planning scheme of the electrochemical energy storage power station is morereasonable and effective.