Energy Storage Trading Method Based on Multi-Market Trading Mechanism

Abstract

The present invention discloses an energy storage trading method based on a multi-market trading mechanism, belonging to the technical field of power energy storage optimization. The method includes the following steps: standardization of multi-market device parameters; evaluation of device health status; prediction and analysis of maintenance costs; comprehensive evaluation of benefits and costs; screening of devices under expected benefits; recommendation of charge and discharge devices; and energy storage trading completion process. By restricting users to select charge and discharge devices according to the expected health status index, the present invention optimizes the charge and discharge frequency of the charge and discharge devices, makes the charge and discharge devices operate more stably, thereby reducing the interruption of market participation caused by the failure of the charge and discharge devices and lowering the maintenance cost, and solves the problem in the prior art that the frequent charge and discharge leads to the aggravation of the loss of the charge and discharge devices and thus the increase of the maintenance cost.

Description

Technical Field

[0001] The present invention relates to the technical field of power energy storage optimization, and particularly to an energy storage trading method based on a multi-market trading mechanism. Background Art

[0002] With the rapid development of renewable energy (such as wind energy and solar energy), the power system is facing the problems of uncertainty and volatility in energy supply and demand. Energy storage technology can balance energy supply and demand and improve the stability and flexibility of the power system by storing electricity and releasing it during peak demand periods. However, since a single market often fails to fully realize the economic benefits of energy storage devices, multi-market trading of energy storage (such as spot markets, ancillary service markets, capacity markets, etc.) has gradually become an effective way to solve this problem. Under this multi-market trading mechanism, energy storage can participate in trading in different markets, earn profits, and at the same time provide various services for the power grid, which conforms to the trend of the energy market gradually moving towards diversification and flexibility.

[0003] Currently, the multi-market energy storage trading mechanism mainly includes: the charging and discharging equipment in the energy storage system dynamically executes charging, discharging, or standby operations according to the price signals and dispatching requirements of multiple markets; by precisely regulating the power output and input of the battery, the charging and discharging equipment provides peak shaving, frequency modulation, and reserve capacity services for the power grid, while complying with market rules and technical constraints to maximize profits.

[0004] For example, a method for planning multiple energy supply entities in an integrated energy market under a carbon trading mechanism disclosed in the invention patent with the publication number of CN113344651A includes: establishing a calculation model for the carbon trading cost of multiple energy supply entities; establishing a Nash equilibrium model for the integrated energy market under the carbon trading mechanism; based on the calculation model for the carbon trading cost of multiple energy supply entities and the Nash equilibrium model for the integrated energy market under the carbon trading mechanism, constructing a two-layer optimization model for the equilibrium bidding of multiple energy supply entities; designing a two-layer improved differential evolution algorithm to solve the two-layer optimization model for the equilibrium bidding of multiple energy supply entities.

[0005] For example, an optimized operation method, device, and equipment for an energy storage power station participating in multiple types of power markets disclosed in the invention patent with the publication number of CN117094849A includes: obtaining the trading data and performance parameters of the energy storage power station, the operation parameters of the power market, and the electricity price data; calculating the planned charging and discharging power of the energy storage power station for the day ahead; according to the planned charging and discharging power of the energy storage power station for the day ahead and the preset day-ahead electricity price prediction scenario, considering the aging cost to construct a joint optimization trading model for the energy storage power station to participate in two types of day-ahead markets, and obtaining the joint optimized operation plan for the day-ahead market of the energy storage power station. The joint optimized operation plan for the day-ahead market of the energy storage power station includes the planned charging and discharging power of the energy storage power station within the day and the planned reserve capacity within the day.

[0006] However, in the process of implementing the inventive technical solution in the embodiments of the present application, it is found that the above technologies have at least the following technical problems:

[0007] In the prior art, frequent high-power charge and discharge operations will accelerate the loss of battery chemical materials and increase the risk of thermal runaway; power electronic devices (such as inverters, IGBT modules) are subjected to voltage and current shocks during frequent switching, resulting in an accelerated aging rate; mechanical devices (such as switches and contactors) are severely worn due to excessive start-stop frequencies. Therefore, the prior art has the problem that frequent charge and discharge lead to increased loss of charge and discharge equipment, and thus increased maintenance costs. Summary of the Invention

[0008] The embodiments of the present application provide an energy storage trading method based on a multi-market trading mechanism, which solves the problem in the prior art that frequent charge and discharge lead to increased loss of charge and discharge equipment and thus increased maintenance costs, and realizes restricting users to select charge and discharge equipment according to the expected health status index to optimize the charge and discharge frequency of the charge and discharge equipment, making the charge and discharge equipment operate more stably, thereby reducing the interruption of market participation caused by the failure of the charge and discharge equipment and reducing the maintenance costs.

[0009] The embodiments of the present application provide an energy storage trading method based on a multi-market trading mechanism, including the following steps: obtaining the device parameter data of all charge and discharge equipment in multiple markets, and performing preprocessing to output a preprocessed device parameter data set; constructing a device health assessment model according to the preprocessed device parameter data set, obtaining the current device parameter data of the charge and discharge equipment, inputting it into the device health assessment model, and outputting the current health status index corresponding to each charge and discharge equipment, where the current health status index is used to reflect the current health status of the charge and discharge equipment; obtaining the multi-market historical maintenance data, establishing a device maintenance rule model, and combining the current health status index to output a maintenance cost prediction value; obtaining the multi-market historical revenue data, constructing a revenue-maintenance cost joint model, inputting the maintenance cost prediction value into the revenue-maintenance cost joint model, and outputting a revenue prediction value; obtaining the expected revenue value set in the database, extracting the charge and discharge equipment that meets the expected revenue value, and analyzing the recommended usage index for the extracted charge and discharge equipment, where the recommended usage index is used to reflect the intensity of the current recommendation for the use of the charge and discharge equipment; generating a device recommendation list according to the recommended usage index, obtaining the charge and discharge demand data of the user, and matching the charge and discharge equipment from the device recommendation list; when the user performs charge and discharge through the matched charge and discharge equipment, the energy storage trading is completed.

[0010] Further, the specific construction method of the device health assessment model is as follows: Obtain device parameter data at regular intervals and obtain a preprocessed device parameter data set. The device parameter data set includes battery type, battery rated capacity, battery capacity change value, battery nominal voltage, Coulomb efficiency, AC internal resistance, DC internal resistance, and number of abnormal events. Obtain the weights of each piece of data in the device parameter data set for the health status index according to the objective weighting method. Classify the device parameter data set according to the corresponding timestamps. Construct a device health assessment model based on the device parameter data set classified by timestamp and its weights for the health status index.

[0011] Further, the device health assessment model is as follows:

[0012]

[0013] In the formula, HSI t is the health status index at timestamp t, t is the timestamp, e is the natural constant, b is the number of the battery type, b = 1, 2,..., B, and B is the total number of battery type numbers. is the battery rated capacity of the b-th battery, ΔC t is the battery capacity change value at timestamp t. is the AC internal resistance of the b-th battery at timestamp t. is the DC internal resistance of the b-th battery at timestamp t. is the number of abnormal events that occurred to the b-th battery at timestamp t, η b is the Coulomb efficiency of the b-th battery. is the average voltage during discharge of the b-th battery at timestamp t. is the average voltage during charging of the b-th battery at timestamp t. is the battery nominal voltage of the b-th battery, α1 is the capacity change ratio For HSI t the weight, α2 is For HSI t the weight, α3 is For HSI t the weight, α4 is For HSI t the weight.

[0014] Further, the specific construction method of the equipment maintenance rule model is as follows: Obtain multi-market historical maintenance data, where the multi-market historical maintenance data includes: the current health status index of the charging and discharging equipment and the maintenance cost data at each maintenance; standardize the multi-market historical maintenance data; divide the multi-market historical maintenance data into a training set and a test set; establish a preliminary maintenance cost prediction model based on the training set using a linear regression algorithm, and test the preliminary maintenance cost prediction model through the test set to obtain the maintenance cost prediction model.

[0015] Further, the steps for obtaining the revenue-maintenance cost joint model are as follows: Obtain multi-market historical revenue data and multi-market historical maintenance data, and perform preprocessing on them; match and correspond the preprocessed multi-market historical revenue data and multi-market historical maintenance data one by one according to the time stamp; divide the matched and corresponding multi-market historical revenue data and multi-market historical maintenance data into a training set and a test set; train a preliminary revenue-maintenance cost joint model based on the training set using a support vector machine, and test the preliminary revenue-maintenance cost joint model using the test set to obtain the revenue-maintenance cost joint model.

[0016] Further, the specific steps for extracting the charging and discharging equipment that meets the expected revenue value are as follows: Obtain the expected revenue value set in the database, perform a certain number of searches through the gradient descent method to obtain the corresponding expected health status index; obtain the current health status index corresponding to each charging and discharging equipment, and compare the current health status index with the expected health status index in turn to determine whether each charging and discharging equipment needs to be evaluated for the recommended usage index; if the current health status index is lower than the expected health status index, do not evaluate the recommended usage index for this charging and discharging equipment; if the current health status index is not lower than the expected health status index, evaluate the recommended usage index for this charging and discharging equipment.

[0017] Further, the specific analysis process of the recommended usage index is as follows: Obtain the current health status index of the extracted charging and discharging equipment, input the current health status index into the equipment maintenance rule model to obtain the corresponding maintenance cost prediction value, input the maintenance cost prediction value into the revenue-maintenance cost joint model to obtain the corresponding revenue prediction value; obtain the weights of the current health status index, maintenance cost prediction value, and revenue prediction value of the charging and discharging equipment for the recommended usage index respectively through the objective weighting method; construct a recommended usage index formula based on the current health status index, maintenance cost prediction value, revenue prediction value, and their respective weights for the recommended usage index; calculate the recommended usage index through the recommended usage index formula.

[0018] Further, the recommended usage index formula is:

[0019]

[0020] wherein, TJI i is the recommended usage index of the i-th charge and discharge device, i is the number of the charge and discharge device, i = 1, 2,..., I, I is the total number of charge and discharge device numbers, e is the natural constant, is the current health status index of the i-th charge and discharge device, is the expected revenue value of the i-th charge and discharge device, is the predicted maintenance cost of the i-th charge and discharge device, and β1 is the weight for TJI i and β2 is the weight for TJI i and β3 is the weight for TJI i respectively.

[0021] Furthermore, the step of matching the charge and discharge device from the device recommendation list includes: obtaining the charge and discharge demand data of the user and preprocessing it; obtaining a matching index according to the preprocessed charge and discharge demand data, where the matching index is used to reflect the suitability of selecting the charge and discharge device in the device recommendation list according to the charge and discharge demand data of the user; reordering the charge and discharge devices in the device recommendation list from high to low according to the matching index, and recommending the sorted matching device list to the user, where the matching device list contains multiple matching charge and discharge devices.

[0022] Furthermore, the obtaining method of the matching index is: extracting the user location information, distance expectation value, power expectation value, and capacity expectation value from the charge and discharge demand data of the user; extracting the device location information, maximum device power, and maximum device capacity from the device parameter data; obtaining the weights of distance, power, and capacity for the matching index through the objective weighting method; obtaining the matching index according to the user location information, distance expectation value, power expectation value, capacity expectation value, device location information, maximum device power, maximum device capacity, and the weights through the matching index formula; the matching index formula is:

[0023]

[0024] wherein, MI i is the matching index of the i-th charge and discharge device, i is the number of the charge and discharge device, i = 1, 2,..., I, I is the total number of charge and discharge device numbers, CU is the user location information, CD i is the device location information of the i-th charge and discharge device, D exp is the distance expectation value, P exp is the power expectation value, is the maximum device power of the i-th charge and discharge device, CA exp is the capacity expectation value, is the maximum device capacity of the i-th charge and discharge device, γ1 is the weight of distance for MI i and γ2 is the weight of power for MI i and γ3 is the weight of capacity for MI i .

[0025] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0026] 1. By selecting charge and discharge devices with a health status index higher than the expected health status index, and then extracting from these charge and discharge devices those that meet the user's needs, the charge and discharge frequencies among various charge and discharge devices are balanced, making the charge and discharge devices operate more stably, thereby reducing the interruption of market participation caused by the failure of charge and discharge devices, lowering the maintenance cost, and effectively solving the problem in the prior art that frequent charge and discharge lead to increased wear and tear of charge and discharge devices and thus increased maintenance cost.

[0027] 2. By comparing the current health status index with the expected health status index, devices that meet the expected revenue value are screened out, thereby achieving precise screening and efficient allocation of charge and discharge devices in multi-market transactions, and enhancing the economy and reliability of energy storage transactions.

[0028] 3. By calculating the matching index based on user demand data and device parameters, the devices that best meet the user's needs are screened out, thereby achieving efficient utilization of energy storage device resources and optimizing the user's trading experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flowchart of an energy storage trading method based on a multi-market trading mechanism provided in the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The embodiments of the present application provide an energy storage trading method based on a multi-market trading mechanism, which solves the problem in the prior art that frequent charge and discharge lead to increased wear and tear of charge and discharge devices and thus increased maintenance cost. By restricting the user's selection of charge and discharge devices according to the expected health status index to optimize the charge and discharge frequency of the charge and discharge devices, the charge and discharge devices operate more stably, thereby reducing the interruption of market participation caused by the failure of charge and discharge devices and lowering the maintenance cost.

[0031] To better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0032] Such as Figure 1As shown in the figure, it is a flowchart of an energy storage trading method based on a multi-market trading mechanism provided by an embodiment of the present application. The method includes the following steps: Standardization of multi-market device parameters: Obtain the device parameter data of all charging and discharging devices in the multi-market, and perform preprocessing to output a preprocessed device parameter data set; where the preprocessing includes data standardization to eliminate data units.

[0033] Device health status assessment: Construct a device health assessment model based on the preprocessed device parameter data set, obtain the current device parameter data of the charging and discharging devices, input it into the device health assessment model, and output the current health status index corresponding to each charging and discharging device. The current health status index is used to reflect the current health status of the charging and discharging devices.

[0034] Maintenance cost prediction and analysis: Obtain the multi-market historical maintenance data, establish a device maintenance rule model, and combine it with the current health status index to output the maintenance cost prediction value.

[0035] Comprehensive assessment of revenue and cost: Obtain the multi-market historical revenue data, construct a revenue-maintenance cost joint model, input the maintenance cost prediction value into the revenue-maintenance cost joint model, and output the expected revenue value.

[0036] Device screening under expected revenue: Obtain the expected revenue value set by the database, extract the charging and discharging devices that meet the expected revenue value, and analyze the recommended usage index for the extracted charging and discharging devices. The recommended usage index is used to reflect the intensity of the current recommendation for using the charging and discharging devices.

[0037] Recommendation of charging and discharging devices: Generate a device recommendation list according to the recommended usage index, obtain the charging and discharging demand data of the user, and match the charging and discharging devices from the device recommendation list.

[0038] Energy storage trading completion process: When the user charges and discharges through the matched charging and discharging devices, the energy storage trading is completed.

[0039] Furthermore, the specific construction method of the device health assessment model is as follows: Obtain the device parameter data at regular intervals and obtain the preprocessed device parameter data set. The device parameter data set includes battery type, battery rated capacity, battery capacity change value, battery nominal voltage, Coulomb efficiency, AC internal resistance, DC internal resistance, and the number of abnormal events; Obtain the weights of each data in the device parameter data set for the health status index according to the objective weighting method; Classify the device parameter data set according to the corresponding time stamps; Construct a device health assessment model according to the device parameter data set classified by time stamps and its weights for the health status index.

[0040] Among them, establishing an equipment health assessment model can significantly improve the scientific and intelligent level of energy storage equipment management. By dynamically calculating the health status index of charging and discharging equipment, operators can timely discover potential problems and take measures, thereby reducing equipment failure rate and maintenance costs.

[0041] In this embodiment, the battery types include lithium-ion batteries, such as NMC (nickel manganese cobalt), LFP (lithium iron phosphate), etc.; in large-scale energy storage systems, lithium iron phosphate (LFP) is more popular due to its high safety and long cycle life.

[0042] The unit of battery rated capacity is MWh (megawatt-hour). The battery rated capacity of large-scale energy storage power stations can range from dozens of megawatt-hours (MWh) to hundreds of megawatt-hours. For example, a medium-sized energy storage power station may be equipped with a battery capacity of 50 MWh.

[0043] The battery capacity change value means that after a new battery is used for a period of time, its capacity may decrease. The average annual capacity attenuation rate is about 2%-3%. Assuming that the capacity remains above 80% after 10 years, the specific change value depends on maintenance and usage conditions; assuming that after one year, the capacity decreases from 54 MWh to 52 MWh, then the battery capacity change value is 2 MWh.

[0044] The battery nominal voltage refers to the total nominal voltage. The nominal voltage of a single battery cell is 3.2V to 3.7V (taking LFP and NMC as examples), and the total nominal voltage of the battery pack depends on the number of series connections and may be between several hundred volts and over a thousand volts.

[0045] The Coulomb efficiency represents the energy conversion efficiency during the charging and discharging processes. The Coulomb efficiency of high-quality energy storage systems is usually between 95% and 98%.

[0046] For large-scale energy storage systems, due to the use of a large number of battery modules in parallel or series, the AC internal resistance can be as low as a few milliohms (mΩ) to dozens of milliohms. For example, the DC internal resistance of a well-maintained lithium iron phosphate (LFP) battery module may be below 10 mΩ. The DC internal resistance can be measured by the large current pulse discharge method, that is, applying a short-term large current across the battery and measuring the voltage drop, and then obtaining the DC internal resistance by dividing the voltage difference by the current.

[0047] The number of abnormal events includes the number of software failures, thermal management problems, or abnormal battery performance. Generally, the number of abnormal events corresponding to a well-maintained energy storage power station can be controlled at a very low level.

[0048] Furthermore, the equipment health assessment model is:

[0049]

[0050] In the formula, HSI tThe health state index at time stamp t, where t is the time stamp, e is the natural constant, b is the number of the battery type, b = 1, 2,..., B, and B is the total number of battery type numbers. is the rated battery capacity of the b-th battery, ΔC t is the battery capacity change value at time stamp t. is the AC internal resistance of the b-th battery at time stamp t. is the DC internal resistance of the b-th battery at time stamp t. is the number of abnormal events that occurred to the b-th battery at time stamp t, η b is the Coulomb efficiency of the b-th battery. is the average voltage during discharge of the b-th battery at time stamp t. is the average voltage during charging of the b-th battery at time stamp t. is the nominal battery voltage of the b-th battery, and α1 is the capacity change ratio. For HSI t the weight, and α2 is For HSI t the weight, and α3 is For HSI t the weight, and α4 is For HSI t the weight.

[0051] In this embodiment, ΔC t = 2 MWh. η b = 0.96. α1 = 0.3, α2 = 0.25, α3 = 0.25, α4 = 0.2, and the resulting health state index is 0.0016. The lower the health state index, the worse the health state of the charging and discharging device, and the higher the health state index, the better the health state of the charging and discharging device.

[0052] Furthermore, the specific construction method of the equipment maintenance rule model is as follows: Obtain multi-market historical maintenance data, where the multi-market historical maintenance data includes: the current health state index and maintenance cost data of the charging and discharging device each time maintenance is performed; standardize the multi-market historical maintenance data; divide the multi-market historical maintenance data into a training set and a test set; establish a preliminary maintenance cost prediction model based on the training set using a linear regression algorithm, and test the preliminary maintenance cost prediction model through the test set to obtain the maintenance cost prediction model.

[0053] In this embodiment, the equipment maintenance law model aims to help operators or users reasonably plan equipment maintenance schedules and avoid over-maintenance or under-maintenance. Traditional energy storage equipment maintenance is usually based on fixed time intervals or experience. However, this model provides a more targeted and scientific prediction solution by analyzing the historical correlation between the equipment health status and maintenance costs, thereby effectively reducing the operation and maintenance costs of the equipment and extending its service life. By introducing the equipment health status index into the maintenance cost prediction process, equipment maintenance has been transformed from passive response to active optimization. By introducing the linear regression algorithm, the model can dynamically adjust the prediction results to adapt to different market scenarios and equipment usage conditions. By applying the equipment maintenance law model, operators can predict the future maintenance requirements and corresponding costs of the equipment in real time, thus achieving refined management. Compared with traditional methods, this model improves the prediction accuracy while reducing unnecessary maintenance times and resource waste. In addition, this model also provides more reliable maintenance cost inputs for revenue analysis, improving the overall efficiency of energy storage transactions.

[0054] Further, the steps for obtaining the revenue-maintenance cost joint model are as follows: Obtain multi-market historical revenue data and multi-market historical maintenance data and preprocess them; Match the preprocessed multi-market historical revenue data and multi-market historical maintenance data one by one according to the timestamp; Divide the matched multi-market historical revenue data and multi-market historical maintenance data into a training set and a test set; Based on the support vector machine, train a preliminary revenue-maintenance cost joint model according to the training set, and use the test set to test the preliminary revenue-maintenance cost joint model to obtain the revenue-maintenance cost joint model.

[0055] In this embodiment, the revenue-maintenance cost joint model realizes the dynamic balance analysis of the equipment economic revenue and operation cost. The economic benefit analysis of traditional energy storage equipment often ignores the impact of maintenance costs, resulting in inaccurate revenue evaluation. However, the revenue-maintenance cost joint model combines the maximization of revenue and the minimization of costs by considering both revenue and maintenance costs simultaneously, providing more comprehensive equipment usage suggestions. Especially in multi-market scenarios, the revenue-maintenance cost joint model can efficiently handle complex data and provide decision support for market collaboration. By predicting the expected revenue value, operators can better screen and recommend equipment and plan resource allocation.

[0056] The revenue-maintenance cost joint model adopts the support vector machine algorithm, overcoming the deficiency of traditional linear regression methods in dealing with high-dimensional non-linear data. This enables the revenue-maintenance cost joint model to maintain high fitting ability when facing multi-market multi-dimensional data. In addition, the revenue-maintenance cost joint model matches revenue and maintenance data through timestamps, ensuring the time consistency and logical integrity of the analysis results, and significantly improving the reliability of model prediction.

[0057] Further, the specific steps for extracting the charging and discharging equipment that meets the expected revenue value are as follows: Obtain the expected revenue value set in the database, perform a certain number of searches through the gradient descent method to obtain the corresponding expected health status index; obtain the current health status index corresponding to each charging and discharging equipment, compare the current health status index with the expected health status index in sequence, and determine whether each charging and discharging equipment needs to be evaluated for the recommended usage index; if the current health status index is lower than the expected health status index, do not evaluate the recommended usage index for this charging and discharging equipment; if the current health status index is not lower than the expected health status index, evaluate the recommended usage index for this charging and discharging equipment.

[0058] In this embodiment, the actual conditions of balancing the user's revenue target and the equipment status are considered to prevent inefficient or overused equipment from entering the recommendation process. By comparing the current health status index with the expected health status index, it can be ensured that the recommended equipment has higher availability and economy, optimizing the equipment selection process. At the same time, this process also provides users with a data-driven basis for selection, significantly reducing decision-making complexity.

[0059] By introducing the gradient descent method to find the expected health status index, the iterative optimization algorithm is used to achieve the quantitative matching of the revenue target and the equipment status. Compared with the traditional experience-based screening method, this method has stronger adaptability and accuracy. In addition, by judging whether the equipment needs to be evaluated for the recommended usage index, the redundancy of calculation and screening is further reduced, improving the efficiency of the overall screening process.

[0060] Further, the specific analysis process of the recommended usage index is as follows: Obtain the current health status index of the extracted charging and discharging equipment, input the current health status index into the equipment maintenance rule model to obtain the corresponding maintenance cost prediction value, input the maintenance cost prediction value into the revenue-maintenance cost joint model to obtain the corresponding revenue prediction value; obtain the weights of the current health status index, maintenance cost prediction value, and revenue prediction value of the charging and discharging equipment for the recommended usage index respectively through the objective weighting method; construct a recommended usage index formula based on the current health status index, maintenance cost prediction value, revenue prediction value, and their respective weights for the recommended usage index; calculate the recommended usage index through the recommended usage index formula.

[0061] In this embodiment, analyzing the recommended usage index of the extracted charging and discharging equipment can provide a scientifically analyzed equipment recommendation level for users or operators, ensuring that equipment with a high usage priority maximizes revenue and reduces cost risks.

[0062] Further, the recommended usage index formula is:

[0063]

[0064] Wherein, TJI i is the recommended usage index of the i-th charge and discharge device, i is the number of the charge and discharge device, i = 1, 2,..., I, and I is the total number of charge and discharge device numbers, is the current health status index of the i-th charge and discharge device, is the expected revenue value of the i-th charge and discharge device, is the predicted maintenance cost value of the i-th charge and discharge device, and β1 is the weight for TJI i β2 is the weight for TJI i β3 is the weight for TJI i the weight.

[0065] In this embodiment, β1 = 0.4, β2 = 0.4, β3 = 0.2, and the recommended usage index is 2.1819. The higher the recommended usage index, the better.

[0066] Furthermore, the steps of matching the charge and discharge device from the device recommendation list include: obtaining the charge and discharge demand data of the user and preprocessing it; obtaining a matching index based on the preprocessed charge and discharge demand data, where the matching index is used to reflect the suitability of selecting the charge and discharge device in the device recommendation list according to the charge and discharge demand data of the user; re - sorting the charge and discharge devices in the device recommendation list in descending order of the matching index, and recommending the sorted matching device list to the user, where the matching device list contains multiple matching charge and discharge devices.

[0067] In this embodiment, the efficiency of energy storage trading and user satisfaction are improved. By quantifying the adaptation degree between user needs and device capabilities through the matching index, it is ensured that the recommended devices can meet the user's usage requirements. Compared with the traditional random recommendation or manual judgment methods, this method significantly improves the accuracy of device selection through precise matching, reducing problems caused by inconsistent device capabilities and user needs. At the same time, the transparency of device sorting provides users with more freedom of choice, enhancing the user experience. Operators can also optimize device allocation through the matching index, maximizing the economic benefits and service capabilities of the energy storage system.

[0068] Furthermore, the way to obtain the matching index is: extracting the user location information, distance expectation value, power expectation value, and capacity expectation value from the user's charge and discharge demand data; extracting the device location information, maximum device power, and maximum device capacity from the device parameter data; obtaining the weights of distance, power, and capacity for the matching index through the objective weighting method; obtaining the matching index through the matching index formula according to the user location information, distance expectation value, power expectation value, capacity expectation value, device location information, maximum device power, maximum device capacity, and the weights.

[0069] Among them, the process of obtaining the matching index realizes the precise matching of user requirements and device capabilities, ensuring that the recommended devices not only meet the functional requirements but also take into account economy and convenience. Thus, users can clearly understand the matching priorities of charging and discharging devices and quickly select devices suitable for their own needs. At the same time, operators can optimize the allocation of device resources through the matching index, reduce invalid transactions, and improve the overall utilization efficiency and market service capabilities of charging and discharging devices.

[0070] The matching index formula is as follows:

[0071]

[0072] In the formula, MI i is the matching index of the i-th charging and discharging device, i is the number of the charging and discharging device, i = 1, 2,..., I, I is the total number of charging and discharging device numbers, CU is the user location information, CD i is the device location information of the i-th charging and discharging device, D exp is the distance expectation value, P exp is the power expectation value, is the maximum device power of the i-th charging and discharging device, CA exp is the capacity expectation value, is the maximum device capacity of the i-th charging and discharging device, γ1 is the weight of distance for MI i is the weight of power for MI i is the weight of capacity for MI i is the weight of capacity for MI.

[0073] In this embodiment, |CU - CD i | = 3.24 km, D exp = 5 km, P exp = 150 kW, CA exp = 0.13 MWh, γ1 = 0.5, γ2 = 0.2, γ3 = 0.3, and the obtained matching index is 0.5959. The higher the matching index, the more in line with the user's needs the charging and discharging device is.

[0074] In summary, in the embodiment of the present application, by restricting users to select charging and discharging devices according to the expected health state index to optimize the charging and discharging frequency of the charging and discharging devices, the charging and discharging devices operate more stably, thereby reducing the interruption of market participation caused by the failure of the charging and discharging devices and reducing the maintenance cost.

[0075] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0076] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0077] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, such that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable devices provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0079] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0080] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. The energy storage trading method based on the multi-market trading mechanism is characterized by: The following steps are involved: Obtain the equipment parameter data of all charging and discharging equipment in multiple markets, perform preprocessing, and output the preprocessed equipment parameter data set; Building an equipment health assessment model based on the preprocessed equipment parameter data set, obtaining the current equipment parameter data of the charging and discharging equipment, inputting it into the equipment health assessment model, and outputting the current health status index corresponding to each charging and discharging equipment, wherein the current health status index is used to reflect the current health status of the charging and discharging equipment; The specific construction method of the equipment health assessment model is: Acquire device parameter data at regular intervals and obtain a preprocessed device parameter data set, wherein the device parameter data set includes battery type, battery rated capacity, battery capacity change value, battery nominal voltage, coulomb efficiency, AC internal resistance, DC internal resistance, and number of abnormal events; Obtain the weight of each data in the equipment parameter data set for the health status index according to the objective weighting method; Classify the device parameter data set according to the corresponding timestamps; Construct a device health assessment model based on the device parameter data set classified by timestamp and its weight to the health status index; The equipment health assessment model is: Where, HSI t is the health status index at timestamp t, t is the timestamp, e is the natural constant, b is the battery type number, b=1,2,...,B, B is the total number of battery type numbers, is the rated capacity of the b-th battery, ΔC t is the battery capacity change at timestamp t, is the AC internal resistance of the b-th battery at time stamp t, is the DC internal resistance of the b-th battery at time stamp t, is the number of abnormal events that occurred in the b-th battery at timestamp t, η b is the coulombic efficiency of the b-th battery, is the average voltage of the b-th battery when it is discharged at time stamp t, is the average voltage of the b-th battery when it is charged at time stamp t, is the battery nominal voltage of the b-th battery, α1 is the capacity change ratio For HSI t The weight of For HSI t The weight of For HSI t The weight of α4 is For HSI t The weight of Obtain historical maintenance data from multiple markets, establish equipment maintenance law models, combine current health status indexes, and output maintenance cost forecasts; The specific construction method of the equipment maintenance law model is: Acquire multi-market historical maintenance data, wherein the multi-market historical maintenance data includes: current health status index and maintenance cost data of the charging and discharging equipment at each maintenance; Standardize historical maintenance data across multiple markets; Divide the multi-market historical maintenance data into training and testing sets; A preliminary maintenance cost prediction model is established based on the training set based on the linear regression algorithm, and the preliminary maintenance cost prediction model is tested through the test set to obtain a maintenance cost prediction model; Obtain historical revenue data from multiple markets, build a revenue-maintenance cost joint model, input maintenance cost forecast values ​​into the revenue-maintenance cost joint model, and output the revenue forecast value; The steps for obtaining the benefit-maintenance cost joint model are: Obtain multi-market historical return data and multi-market historical maintenance data, and pre-process them; Match the pre-processed multi-market historical revenue data and multi-market historical maintenance data one by one according to timestamps; Divide the matched multi-market historical return data and multi-market historical maintenance data into a training set and a test set; Based on the support vector machine, a preliminary benefit-maintenance cost joint model is trained according to the training set, and the preliminary benefit-maintenance cost joint model is tested using the test set to obtain a benefit-maintenance cost joint model; Obtaining the expected benefit value set in the database, extracting the charging and discharging equipment that meets the expected benefit value, and analyzing the recommended use index for the extracted charging and discharging equipment, wherein the recommended use index is used to reflect the intensity of the current recommended use of the charging and discharging equipment; The specific analysis process of the recommended use index is as follows: Obtain the current health status index of the extracted charging and discharging equipment, input the current health status index into the equipment maintenance law model, obtain the corresponding maintenance cost forecast value, input the maintenance cost forecast value into the benefit-maintenance cost joint model, and obtain the corresponding benefit forecast value; Obtain the weights of the current health status index, maintenance cost forecast value, and benefit forecast value of the charging and discharging equipment for the recommended use index through an objective weighting method; Constructing a recommended use index formula based on the current health status index, the maintenance cost forecast value, the benefit forecast value and their respective weights for the recommended use index; The recommended use index is calculated by the recommended use index formula; The recommended index formula is: Where, TJI i is the recommended use index of the i-th charging and discharging equipment, i is the number of the charging and discharging equipment, i=1,2,...,I, I is the total number of charging and discharging equipment numbers, e is a natural constant, is the current health status index of the i-th charging and discharging device, is the estimated value of the revenue of the i-th charging and discharging equipment, is the maintenance cost prediction value of the i-th charging and discharging equipment, β1 is For TJI i The weight of For TJI i The weight of For TJI i The weight of Generate a device recommendation list based on the recommended usage index, obtain the user's charging and discharging demand data, and match the charging and discharging device from the device recommendation list; When the user charges and discharges through the matched charging and discharging equipment, the energy storage transaction is completed.

2. The energy storage trading method based on the multi-market trading mechanism as claimed in claim 1, characterized in that: The specific steps of extracting the charging and discharging equipment that meets the expected benefit value are: Obtain the expected benefit value set in the database, perform a certain number of searches using the gradient descent method, and obtain the corresponding expected health status index; Obtain the current health status index corresponding to each charging and discharging device, compare the current health status index with the expected health status index in turn, and determine whether each charging and discharging device needs to be evaluated for the recommended use index; If the current health status index is lower than the expected health status index, the recommended use index evaluation will not be performed on the charging and discharging equipment; If the current health status index is not lower than the expected health status index, the charging and discharging equipment is evaluated for a recommended usage index.

3. The energy storage trading method based on the multi-market trading mechanism as claimed in claim 1, characterized in that: The step of matching the charging and discharging device from the device recommendation list includes: Obtain the user's charging and discharging demand data and pre-process it; A matching index is obtained based on the preprocessed charging and discharging demand data, where the matching index is used to reflect the suitability of the charging and discharging device in the device recommendation list selected based on the user's charging and discharging demand data; The charging and discharging devices in the device recommendation list are re-sorted from high to low according to the matching index, and the sorted matching device list is recommended to the user, wherein the matching device list includes multiple matching charging and discharging devices.

4. The energy storage trading method based on the multi-market trading mechanism as claimed in claim 3, characterized in that: The matching index is obtained as follows: Extracting user location information, distance expectation value, power expectation value and capacity expectation value from the user's charging and discharging demand data; Extract device location information, maximum device power and maximum device capacity from device parameter data; Obtain the weights of distance, power and capacity for the matching index through objective weighting method; The matching index is obtained by using the matching index formula according to the user location information, expected distance, expected power, expected capacity, device positioning information, maximum device power, maximum device capacity and weight; The matching index formula is: Where, MI i is the matching index of the i-th charging and discharging device, i is the number of the charging and discharging device, i=1,2,...,I, I is the total number of charging and discharging device numbers, CU is the user location information, CD i is the device location information of the i-th charging and discharging device, D exp is the expected value of the distance, P exp is the expected power value, is the maximum power of the i-th charging and discharging device, CA exp is the expected capacity value, is the maximum capacity of the i-th charging and discharging device, γ1 is the distance to MI i The weight, γ2 is the power for MI i The weight of γ3 is the capacity for MI i The weight of .

Images