A grid-type energy storage system and control method thereof
Through JSON formatted data interface and real-time data processing engine, combined with message queues and event response processing, the energy allocation solution is optimized, and the existing system's lack of data processing delay and response speed is solved, and the energy management efficiency of smart grids is improved. Reliability of power supply.
Patent Information
- Application Number
- CN202411141833.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-20
AI Technical Summary
The existing grid-type energy storage systems have delays and insufficient response speeds in processing batch real-time data and efficient energy allocation, making it difficult to cope with the challenges of grid failures or peak demand, affecting the reliability of energy management and power supply.
Using JSON formatted data interface and TCP/IP protocol, combined with Apache Kafka message queue and Spark Streaming real-time data processing engine, dynamically estimate energy demand through state analysis and prediction modules, use Esper engine to perform incident response processing, and optimize energy allocation schemes. And adjust the charging and discharging strategy through the energy scheduling and management module.
It significantly improves the energy management efficiency of smart grids, realizes real-time data processing and predictive resource allocation, reduces energy waste, and improves the reliability and economics of the grid.
Smart Images

Figure CN118676970B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of smart grid technology, and in particular to a grid-connected energy storage system and a control method thereof. Background Art
[0002] The field of smart grid technology focuses on enhancing the management efficiency and reliability of the power grid by integrating digital communications and information technology into the power network. The technology field enables the power grid to automatically monitor the flow of energy and the status of the power system, and to adjust and optimize the supply and consumption of energy in real time. Smart grids support more access to renewable energy, such as wind and solar energy, thereby reducing dependence on fossil fuels. Smart grids can also improve the economic and environmental sustainability of the power grid, provide better power quality and power supply reliability, and enhance the ability to respond to grid faults. By introducing advanced metering facilities, demand response technology and automated management, smart grids aim to achieve a more interactive, automated and efficient power system.
[0003] Among them, the grid-based energy storage system is a highly modular and automated power system designed to optimize power storage and distribution. The system can quickly respond to changes in power demand by exchanging status and demand information between multiple storage units in real time, effectively optimize energy configuration and improve the overall efficiency of the system. This system is particularly suitable for environments that require a high degree of energy management flexibility, such as smart grids equipped with renewable energy generation. Its main uses include improving energy utilization efficiency, ensuring the stable operation of the power grid, and promoting the integration of renewable energy, thereby supporting sustainable energy development strategies.
[0004] The existing grid-connected energy storage system improves the interactivity and automation level of the power system, but it still has limitations in processing batch real-time data and performing efficient energy distribution. Data processing in the existing system relies on traditional batch processing methods, which leads to delays in data processing and makes it difficult to achieve real-time energy management and scheduling. The existing technology also shows shortcomings in event response mechanisms, especially when the power grid fails or demand surges. The reaction speed and resource allocation efficiency of the existing system are still insufficient to cope with the challenges of peak hours. Technical limitations not only increase energy waste, but also affect the stability of the power grid and the reliability of power supply. In smart grid technology, although the access to renewable energy is increased, under highly variable energy input and complex power grid environment, the existing scheduling and monitoring systems are difficult to accurately predict and manage dynamic changes, thus affecting the efficiency and economy of the entire power grid. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a grid-type energy storage system and a control method thereof.
[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A grid-type energy storage system comprises:
[0007] The data interface and transfer module formats the key parameters of the energy storage unit, such as voltage, current, temperature, and remaining capacity, in JSON based on the energy storage unit data. The data is sent to the central control configuration via the TCP / IP protocol to obtain the encoded data stream, which is then input into the Apache Kafka message queue to obtain the queue data stream.
[0008] The state analysis and prediction module inputs the queue data stream into a real-time data processing engine based on Spark Streaming, analyzes the voltage, current, temperature and capacity data, obtains state time series analysis results, and dynamically estimates the energy demand of each unit based on the state time series analysis results to obtain demand prediction results;
[0009] The event response and strategy adjustment module identifies energy demand peaks and sudden batch power consumption events based on the demand forecast results, matches event patterns, uses the Esper engine to perform event processing, obtains event response decisions, uses the event response decisions to adjust energy supply, and obtains an optimized energy allocation plan;
[0010] The energy dispatching and management module receives the optimized energy allocation plan, adjusts the charging and discharging strategy according to demand, records the energy flow status, analyzes the energy storage charging and discharging efficiency and adaptability, and obtains the intelligent grid energy storage charging and discharging dispatching data results.
[0011] As a further solution of the present invention, the step of acquiring the queue data flow is specifically:
[0012] Based on the energy storage unit data, each key parameter value of the energy storage unit is converted into the attributes of the JSON object through the formatting function, using the formula:
[0013] ,
[0014] Generate a formatted data packet, where K represents the key parameter. represents the real-time value of parameter k, f represents the data formatting function, represents the weight of parameter k, N represents the total number of parameters, Indicates the formatted data packet;
[0015] The formatted data packet is sent through the TCP / IP protocol using the sending function, using the formula:
[0016] ,
[0017] Generates an encoded data stream where represents the i-th formatted data packet, n represents the number of data packets, s represents the sending function, Represents the encoded data stream;
[0018] The encoded data stream is input into the Apache Kafka message queue, using the queue function, using the formula:
[0019] ,
[0020] Generate a queue data stream, where represents the encoded data stream, q represents the queue function, L represents the load factor, Represents queue data flow.
[0021] As a further solution of the present invention, the step of obtaining the state time series analysis result is specifically:
[0022] The queue data stream is processed in real time by the Spark Streaming engine, and the transformation function is applied and the processing speed and responsiveness of the data stream are adjusted in combination with the adjustment parameters, using the formula:
[0023] ,
[0024] Generate processed data streams, is the baseline adjustment factor, and k are dynamic adjustment parameters, Calculate the queue length, represents the processed data stream, Represents the input queue data flow, Conversion functions for real-time processing;
[0025] Extract key electrical features from the processed data stream, including voltage, current, temperature, and capacity, and weight them based on feature weights, using the formula:
[0026] ,
[0027] Generate a data stream feature set, where h is the feature extraction function, is the feature weight, Z is the normalization constant, represents the data stream feature set, They are voltage, current, temperature and capacity respectively;
[0028] The data stream feature set is subjected to time series analysis, using the analysis function combined with the time factor, and adopting the formula:
[0029] ,
[0030] Generate state time series analysis results, where t represents the time variable, is the decay constant, is the time influence intensity factor, represents the state time series analysis results, represents the data stream feature set, and m represents the time series analysis function.
[0031] As a further solution of the present invention, the step of obtaining the demand forecast result is specifically as follows:
[0032] The energy data of each unit is integrated from the state time series analysis results, and the historical data is combined with the current data using the enhanced integration function, through the formula:
[0033] ,
[0034] Generate integrated data, where represents the historical energy demand of the ith unit, n is the total number of units, represents the state time series analysis result, f represents the enhanced integration function, Represents the integrated data;
[0035] Apply the forecasting model to the integrated data to dynamically calculate the demand in the future time period, through the formula:
[0036] ,
[0037] Generate forecasted energy demand, where is the time decay constant, t is the time interval from data integration to prediction time, represents the predicted energy demand, represents the integrated data, and p represents the demand forecasting model;
[0038] The predicted energy demand is formatted and output and matched to the energy unit management configuration through the formula:
[0039] ,
[0040] Generate demand forecast results, where format is a formatting function. is the formatting parameter, represents the demand forecast result, Represents predicted energy demand.
[0041] As a further solution of the present invention, the step of obtaining the event response decision is specifically as follows:
[0042] Using the demand forecast results, energy demand peaks are defined and sudden batch power consumption events are identified. A threshold identification method is applied, using the formula:
[0043] ,
[0044] Generate an event pattern where is the dynamically adjusted demand peak threshold, n represents the number of data points in the time series, Represents an event mode, represents the demand forecast result, and x represents the energy demand data at a single time point;
[0045] The event pattern is applied to the Esper engine for pattern matching to identify key energy events, using the formula:
[0046] ,
[0047] Generates matching events, where is the importance weight of each event, is the sum of event weights, Indicates the event after matching, Represents the event pattern, match represents the pattern matching function of the Esper engine;
[0048] Based on the matched events, the decision logic is executed and the identified events are responded to to generate event response decisions, using the formula:
[0049] ,
[0050] Where k is the adjustment factor, m represents the number of matching events, represents an incident response decision, Represents the event after matching, and decision represents the decision function applied to the matching event.
[0051] As a further solution of the present invention, the step of obtaining the optimized energy allocation solution is specifically as follows:
[0052] Extract data from the incident response decision, analyze and identify the energy supply strategy that needs to be adjusted, through the formula:
[0053] ,
[0054] Generate analyzed energy data, where represents the data of the i-th event response decision, is the adjustment factor, represents the analyzed energy data;
[0055] According to the analyzed energy data, the energy supply strategy is adjusted to match the current and predicted energy demand, through the formula:
[0056] ,
[0057] Generate adjusted supply data, where m is the function that adjusts energy supply, represents the adjusted supply data, Indicates the maximum value in the analyzed data. represents the analyzed energy data;
[0058] To convert the adjusted supply data into an operational strategy, the formula is:
[0059] ,
[0060] Form an optimized energy allocation plan, where: is the function that converts supply adjustments into policies, It represents the optimization of energy distribution scheme. represents the adjusted supply data, represents a logarithmic transformation for enhancing small value adjustments, Represents the logarithm base.
[0061] As a further solution of the present invention, the steps for obtaining the smart grid energy storage charging and discharging scheduling data results are specifically as follows:
[0062] The optimized energy allocation plan is received and combined with the real-time energy consumption data to form a summary data view, using the formula:
[0063] ,
[0064] in, Represents aggregated data, represents the i-th strategy result data point, is the corresponding energy consumption data, n is the number of data points;
[0065] The aggregated data view is used to monitor energy flow, match supply with demand, and adjust energy charging and discharging, using the formula:
[0066] ,
[0067] Generate energy flow states, where is the weight factor for each data point, v is the global parameter for adjusting the nonlinear response, Indicates the energy flow state, Represents aggregated data;
[0068] Based on the energy flow state, the efficiency and adaptability of energy distribution are analyzed through the formula:
[0069] ,
[0070] Generate smart grid energy storage charging and discharging scheduling data results, including: is the efficiency benchmark for each state, are the coefficients used for adaptability analysis, It represents the intelligent grid energy storage charging and discharging scheduling data results, Indicates the energy flow state.
[0071] A grid-type energy storage control method is implemented based on the above-mentioned grid-type energy storage system, comprising the following steps:
[0072] S1: Collect the voltage, current, temperature, and remaining capacity data of the energy storage unit through the control interface, format the data into JSON, and control the data transmission process through the TCP / IP protocol, optimize the data packet size and transmission frequency, match the network conditions, and generate a coded data stream;
[0073] S2: using the encoded data stream, the data is fed into Apache Kafka for control processing, and the reliability of message processing is optimized by controlling the ordering and error detection of the data stream in the message queue, thereby obtaining a queue data stream;
[0074] S3: Perform data control analysis on the queue data stream in the Spark Streaming environment, continuously monitor the voltage, current, temperature and capacity, analyze the time series changes of the parameters, control and identify abnormal patterns, and generate state time series analysis results;
[0075] S4: Based on the state time series analysis results, control the calculation of demand forecast, define and identify energy demand peaks and sudden power consumption events, control real-time response processing to events, adjust the output configuration of energy storage units, and obtain event response decisions;
[0076] S5: Use the event response decision to control the energy supply and output of energy storage, adjust the charging and discharging strategy according to demand, monitor and control the energy flow and consumption of the network, analyze and adjust the energy allocation strategy and optimize the energy storage control configuration performance, and establish smart grid energy storage charging and discharging scheduling data results.
[0077] Compared with the prior art, the advantages and positive effects of the present invention are:
[0078] In the present invention, the energy management efficiency in the smart grid is significantly improved through real-time data formatting and efficient information flow. The JSON formatted data interface plus the TCP / IP protocol are used to ensure the real-time transmission and accuracy of the data. The message queue system using Apache Kafka as the center optimizes the data processing process, improves the processing throughput and reduces the response time of the system, so that the power grid can manage and allocate resources more efficiently. The state analysis and demand forecasting performed by Spark Streaming enable the power grid to not only respond in real time, but also predict future changes in energy demand, which helps to adjust resource allocation in advance and reduce energy waste. The event response processing of the Esper engine further enhances the system's adaptability to emergencies, so that the power grid can quickly and effectively adjust energy supply when facing demand peaks or batch power consumption events, thereby improving the reliability and economy of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] Figure 1 is a system flow chart of the present invention;
[0080] Figure 2 It is a flow chart of queue data flow in the present invention;
[0081] Figure 3 It is a flow chart of the state time series analysis results in the present invention;
[0082] Figure 4 It is a flow chart of the demand forecasting results in the present invention;
[0083] Figure 5 A flowchart of event response decision making in the present invention;
[0084] Figure 6 A flow chart of the energy allocation scheme optimized in the present invention;
[0085] Figure 7 It is a flow chart of the smart grid energy storage charging and discharging scheduling data results in the present invention. DETAILED DESCRIPTION
[0086] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0087] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0088] Embodiment 1:
[0089] See also Figure 1 The present invention provides a technical solution: a grid-type energy storage system comprising:
[0090] The data interface and transfer module formats the key parameters of the energy storage unit, such as voltage, current, temperature, and remaining capacity, in JSON based on the energy storage unit data. The data is sent to the central control configuration via the TCP / IP protocol to obtain the encoded data stream, which is then input into the Apache Kafka message queue to obtain the queue data stream.
[0091] The state analysis and prediction module inputs the queue data stream into the real-time data processing engine based on Spark Streaming, analyzes the voltage, current, temperature and capacity data, obtains the state time series analysis results, and dynamically estimates the energy demand of each unit based on the state time series analysis results to obtain the demand forecast results;
[0092] The event response and strategy adjustment module identifies energy demand peaks and sudden batch power consumption events based on demand forecast results, matches event patterns, uses the Esper engine to perform event processing, obtains event response decisions, uses event response decisions to adjust energy supply, and obtains optimized energy allocation plans;
[0093] The energy dispatching and management module receives the optimized energy distribution plan, adjusts the charging and discharging strategy according to demand, records the energy flow status, analyzes the energy storage charging and discharging efficiency and adaptability, and obtains the intelligent grid energy storage charging and discharging dispatching data results.
[0094] The encoded data stream includes data serialization, transmission optimization, and protocol compatibility. The queue data stream includes message queue length, caching strategy, and transmission stability. The state time series analysis results include voltage trend analysis, current pattern recognition, and temperature anomaly detection. The demand forecasting results include consumption estimation, demand responsiveness, and forecast accuracy. Event response decisions include strategy formulation speed, event triggering accuracy, and decision execution efficiency. The optimized energy allocation plan includes supply adjustment range, output balance, and system response speed. The smart grid energy storage charging and discharging scheduling data results include monitoring system efficiency, data reporting accuracy, and scheduling adaptability.
[0095] See also Figure 2 , the specific steps for obtaining the queue data flow are:
[0096] Based on the energy storage unit data, each key parameter value of the energy storage unit is converted into the attributes of the JSON object through the formatting function, using the formula:
[0097] ,
[0098] Generate a formatted data packet, where K represents the key parameter. represents the real-time value of parameter k, f represents the data formatting function, represents the weight of parameter k, N represents the total number of parameters, Indicates the formatted data packet;
[0099] Use the send function to send the formatted data packet through the TCP / IP protocol, using the formula:
[0100] ,
[0101] Generates an encoded data stream where represents the i-th formatted data packet, n represents the number of data packets, s represents the sending function, Represents the encoded data stream;
[0102] Input the encoded data stream into the Apache Kafka message queue, using the queue function, using the formula:
[0103] ,
[0104] Generate a queue data stream, where represents the encoded data stream, q represents the queue function, L represents the load factor, Represents queue data flow.
[0105] The formatted data packet formula is:
[0106] ,
[0107] K: A collection of all key parameters, such as voltage , Current ,temperature and remaining capacity ;
[0108] : The real-time value of each parameter, such as , , , ;
[0109] : The weight of each parameter, here we assume that the weights are equal and each item is 1;
[0110] N: total number of parameters, here is 4;
[0111] Data preparation: , , , , , ;
[0112] calculate: :
[0113] ,
[0114] This result (10.3625) represents the weighted average parameter value and is used for further data processing.
[0115] The encoded data flow formula is:
[0116] ,
[0117] : The formatted data packet, in the above example it is 10.3625;
[0118] n: number of data packets, assuming there are 3 data packets;
[0119] Assumptions: , n=3;
[0120] calculate :
[0121] ,
[0122] Result Interpretation This result (17.943) represents the processed weighted data stream, which is suitable for network transmission.
[0123] Queue data flow formula:
[0124] ,
[0125] : The encoded data stream, in the above example it is 17.943;
[0126] L: system load factor, assumed to be 5;
[0127] Assumptions: , L=5;
[0128] calculate :
[0129] ,
[0130] This result (21.532) represents the processed data stream that is input to Apache Kafka for subsequent message queue processing.
[0131] See also Figure 3 , the specific steps for obtaining the state time series analysis results are:
[0132] The queue data stream is processed in real time through the Spark Streaming engine, and the transformation function is applied and combined with the adjustment parameters to adjust the processing speed and responsiveness of the data stream. The formula is used:
[0133] ,
[0134] Generate processed data streams, is the baseline adjustment factor, and k are dynamic adjustment parameters, Calculate the queue length, represents the processed data stream, Represents the input queue data flow, Conversion functions for real-time processing;
[0135] Extract key electrical features from the processed data stream, including voltage, current, temperature, and capacity, and weight them based on feature weights, using the formula:
[0136] ,
[0137] Generate a data stream feature set, where h is the feature extraction function, is the feature weight, Z is the normalization constant, represents the data stream feature set, They are voltage, current, temperature and capacity respectively;
[0138] Perform time series analysis on the data stream feature set, using the analysis function and combining the time factor, using the formula:
[0139] ,
[0140] Generate state time series analysis results, where t represents the time variable, is the decay constant, is the time influence intensity factor, represents the state time series analysis results, represents the data stream feature set, and m represents the time series analysis function.
[0141] The processed data flow formula is:
[0142] ,
[0143] : Indicates the input queue data flow Real-time processing conversion functions of applications;
[0144] : Baseline adjustment factor, used to control the basic rate of processing;
[0145] : Dynamic adjustment parameters, used to adjust the response as the length of the data stream changes;
[0146] k: coefficient for adjusting sensitivity, affecting Application intensity;
[0147] : The length of , that is, the number of data items in the data stream;
[0148] Assumptions: Item data, (represents the base value of the processed data), , , k=0.1;
[0149] calculate ;
[0150] Calculate reconciliation ;
[0151] Calculate the total adjustment factor ;
[0152] Calculate the final result ;
[0153] The final processed data flow It means that the adjusted processing intensity is twice of a certain value, which means that the processing engine has a strong responsiveness to large data traffic.
[0154] Data stream feature set formula:
[0155] ,
[0156] : Feature extraction function for The data items k in (voltage V, current I, temperature T and capacity C);
[0157] : The weight of each feature;
[0158] Z: normalization constant to ensure reasonable weight distribution;
[0159] Assumptions: , , , , , , , , ;
[0160] Compute the weighted feature sum:
[0161] ,
[0162] calculate :
[0163] ,
[0164] The final feature set It represents the weighted and normalized result after comprehensively considering all features, which is used for the next step of time series analysis.
[0165] State time series analysis result formula:
[0166] ,
[0167] : Time series analysis functions, processing ;
[0168] t: current time point;
[0169] : time decay constant;
[0170] : Time influence intensity factor;
[0171] Assumptions: (Assume that the base value for time series analysis is three times), t=5, , ;
[0172] Calculate the time factor:
[0173] ,
[0174] Calculate the final result :
[0175] ,
[0176] Final state time series analysis results It means that under the influence of a given time point, the analysis results show that the influence of the feature is weakened, reflecting the significant impact of the time factor on the results.
[0177] See also Figure 4 , the specific steps for obtaining demand forecast results are:
[0178] Integrate the energy data of each unit from the state time series analysis results, and use the enhanced integration function to combine historical data with current data through the formula:
[0179] ,
[0180] Generate integrated data, where represents the historical energy demand of the ith unit, n is the total number of units, represents the state time series analysis result, f represents the enhanced integration function, Represents the integrated data;
[0181] Apply the forecasting model to the integrated data to dynamically calculate the demand in the future time period, through the formula:
[0182] ,
[0183] Generate forecasted energy demand, where is the time decay constant, t is the time interval from data integration to prediction time, represents the predicted energy demand, represents the integrated data, and p represents the demand forecasting model;
[0184] The predicted energy demand is formatted and output and matched to the energy unit management configuration through the formula:
[0185] ,
[0186] Generate demand forecast results, where format is a formatting function. is the formatting parameter, represents the demand forecast result, Represents predicted energy demand.
[0187] The integrated data formula is:
[0188] ,
[0189] : The result of the state time series analysis, assumed to be 1000, represents the energy state index at a certain time point;
[0190] : The integration function can be assumed to be , which means that the state time series results are scaled to be combined with historical data;
[0191] : The historical energy demand of the i-th unit. Assume that the energy demand of the three units is , , ;
[0192] n: total number of units, here is 3;
[0193] calculate :
[0194] ,
[0195] Calculate the average energy demand squared:
[0196] ,
[0197] ,
[0198] Final integration of data :
[0199] ,
[0200] This combined data represents a comprehensive energy demand forecast that takes into account historical energy demand and current conditions.
[0201] Energy demand forecast formula:
[0202] ,
[0203] : Demand forecasting model function, assumed to be , which means converting the aggregated data into specific energy demand forecasts;
[0204] : time decay constant, assumed to be 5;
[0205] t: time interval, from data integration to prediction time, assumed to be 10 hours;
[0206] calculate :
[0207] ,
[0208] Final forecast demand :
[0209] ,
[0210] This forecast demand represents the predicted energy demand after accounting for time decay.
[0211] Demand forecast result formula:
[0212] ,
[0213] : Formatting function, assuming it is simplified to , no format changes;
[0214] : Formatting parameter, affecting the output format, assumed to be 10;
[0215] Formatting forecast requirements:
[0216] ,
[0217] Calculate the final output:
[0218] ,
[0219] The result represents a formatted energy demand forecast ready for reporting to the management system.
[0220] See also Figure 5 , the specific steps for obtaining event response decisions are:
[0221] Using the demand forecast results, define the energy demand peak and identify the sudden batch power consumption events, apply the threshold identification method, through the formula:
[0222] ,
[0223] Generate an event pattern where is the dynamically adjusted demand peak threshold, n represents the number of data points in the time series, Represents an event mode, represents the demand forecast result, and x represents the energy demand data at a single time point;
[0224] Apply the event pattern to the Esper engine for pattern matching and identify key energy events through the formula:
[0225] ,
[0226] Generates matching events, where is the importance weight of each event, is the sum of event weights, Indicates the event after matching, Represents the event pattern, match represents the pattern matching function of the Esper engine;
[0227] Based on the matched events, the decision logic is executed and the identified events are responded to to generate event response decisions, using the formula:
[0228] ,
[0229] Where k is the adjustment factor, m represents the number of matching events, represents an incident response decision, Represents the event after matching, and decision represents the decision function applied to the matching event.
[0230] Event pattern formula:
[0231] ,
[0232] : Energy demand forecast data for every 10 minutes within an hour, in megawatts (MW), ;
[0233] : The set demand peak threshold is 20MW;
[0234] n: The number of data points is 6 (i.e. six ten-minute intervals in one hour);
[0235] Calculate the dynamic threshold adjustment factor: ,
[0236] Then calculate the new threshold: ,
[0237] Filter out eligible energy demands based on the new thresholds:
[0238] ,
[0239] result It indicates that during the observation period, there were three time points when the energy demand exceeded the adjusted peak threshold, which affected the stability and operation of the energy system.
[0240] Event pattern formula:
[0241] ,
[0242] Assumptions Simply returns the number of matched events, here 3 (three events exceeding the threshold);
[0243] : The weight of each event is assumed to be 1;
[0244] (the sum of weights of all events);
[0245] Calculate the weight ratio:
[0246] ,
[0247] Calculate the final result:
[0248] ,
[0249] result It indicates that after weight adjustment, the system identifies critical events with an average impact unit of 1.5, which helps system decision makers to further analyze and respond.
[0250] Incident response decision formula:
[0251] ,
[0252] Assumptions Returns the basic weight of the event decision, which is 2 here (assuming two strategies need to be dealt with);
[0253] k=0.5, m=1.5 (matching events obtained from the previous step);
[0254] Calculate the exponential response:
[0255] ,
[0256] Calculate the final result:
[0257] ,
[0258] result It means that based on the decision weight of the current state of the system, approximately 1.358 units of strategy response strength are required to handle the identified events after adjustment.
[0259] See also Figure 6 ,The specific steps for obtaining the optimized energy allocation plan are:
[0260] Extract data from incident response decisions, analyze and identify energy supply strategies that need to be adjusted, using the formula:
[0261] ,
[0262] Generate analyzed energy data, where represents the data of the i-th event response decision, is the adjustment factor, represents the analyzed energy data;
[0263] According to the analyzed energy data, the energy supply strategy is adjusted to match the current and predicted energy demand through the formula:
[0264] ,
[0265] Generate adjusted supply data, where m is the function that adjusts energy supply, represents the adjusted supply data, Indicates the maximum value in the analyzed data. represents the analyzed energy data;
[0266] To convert the adjusted supply data into an operational strategy, use the formula:
[0267] ,
[0268] Form an optimized energy allocation plan, where: is the function that converts supply adjustments into policies, It represents the optimization of energy distribution scheme. represents the adjusted supply data, represents a logarithmic transformation for enhancing small value adjustments, Represents the logarithm base.
[0269] The energy data formula after analysis:
[0270] ,
[0271] : The value of the i-th decision data point, assumed to be the urgency score of the response;
[0272] : The adjustment coefficient of the i-th data point, reflecting the relative importance of the data point;
[0273] n: the total number of data points, indicating how many decision points are considered;
[0274] Assume there are 3 decision data points, that is, n=3, decision data points 10, 15 and 20 respectively, adjustment factor They are 1.2, 0.8 and 1.0 respectively;
[0275] Calculate the weighted value for each decision point:
[0276] ,
[0277] Sum all weighted values:
[0278] ,
[0279] Divide by :
[0280] ,
[0281] calculate :
[0282] ,
[0283] Income , which represents the weighted average decision urgency.
[0284] Adjusted supply data formula:
[0285] ,
[0286] : Energy adjustment function, assumed to be ;
[0287] : Analyzed energy data;
[0288] : The maximum value of is assumed to be 10;
[0289] use (calculated from the previous step);
[0290] calculate :
[0291] ,
[0292] Compute the sine part:
[0293] ,
[0294] Final result:
[0295] ,
[0296] Income , represents the adjusted energy supply index.
[0297] Formula for optimizing energy allocation scheme:
[0298] ,
[0299] : A function converted into a specific strategy, assuming ;
[0300] : Adjusted supply data;
[0301] : The base of natural logarithm is 2, which is approximately equal to 0.693;
[0302] use (calculated from the previous step):
[0303] Calculate the strategy switching function:
[0304] ,
[0305] Compute the logarithmic part:
[0306] ,
[0307] Final result:
[0308] ,
[0309] Income , representing the comprehensive effect index of the final adjustment strategy.
[0310] See also Figure 7 , the specific steps for obtaining the smart grid energy storage charging and discharging scheduling data results are as follows:
[0311] Receive optimized energy allocation plans and combine them with real-time energy consumption data to form a summary data view using the formula:
[0312] ,
[0313] in, Represents aggregated data, represents the i-th strategy result data point, is the corresponding energy consumption data, n is the number of data points;
[0314] Use the aggregated data view to monitor energy flows, match supply with demand, and adjust energy charging and discharging, using the formula:
[0315] ,
[0316] Generate energy flow states, where is the weight factor for each data point, v is the global parameter for adjusting the nonlinear response, Indicates the energy flow state, Represents aggregated data;
[0317] Based on the energy flow status, the efficiency and adaptability of energy distribution are analyzed through the formula:
[0318] ,
[0319] Generate smart grid energy storage charging and discharging scheduling data results, including: is the efficiency benchmark for each state, are the coefficients used for adaptability analysis, It represents the intelligent grid energy storage charging and discharging scheduling data results, Indicates the energy flow state.
[0320] Summarize data formula:
[0321] ,
[0322] : represents the data point of the i-th optimized energy allocation scheme, assuming that the data comes from the calculation of the previous step or real-time system feedback;
[0323] : represents the i-th energy consumption data point, based on the feedback from the real-time monitoring system;
[0324] n: represents the total number of data points, used for normalized data processing;
[0325] Assume there are 3 strategy result data points: , the corresponding energy consumption data is , then n=3;
[0326] calculate :
[0327] ,
[0328] This result (952.81) represents the integrated data indicator given the policy adjustments and energy consumption data.
[0329] Energy flow state formula:
[0330] ,
[0331] : The integrated data points obtained from the first step;
[0332] : The weight factor for each data point, which is assumed to be the standard weight obtained from historical data or expert system;
[0333] v: global parameter, used to adjust nonlinear response;
[0334] Continuing with the previous example ,set up , v = 2;
[0335] calculate :
[0336] ,
[0337] This result (65.48) represents an indicator of the adjusted state of energy flows.
[0338] Smart grid energy storage charging and discharging scheduling data result formula:
[0339] ,
[0340] : Energy flow status data points obtained from the previous step;
[0341] : The efficiency benchmark for each state, assuming this is based on industry standards or past performance evaluations;
[0342] : Global adaptability coefficient, enhancing the depth of system adaptability analysis;
[0343] use , assuming , ;
[0344] calculate :
[0345] ,
[0346] This result (72.028) represents a comprehensive evaluation index of energy allocation efficiency and system adaptability.
[0347] A grid-type energy storage control method is implemented based on the above-mentioned grid-type energy storage system, and comprises the following steps:
[0348] S1: Collect the voltage, current, temperature, and remaining capacity data of the energy storage unit through the control interface, format the data into JSON, and control the data transmission process through the TCP / IP protocol, optimize the data packet size and transmission frequency, match the network conditions, and generate a coded data stream;
[0349] S2: Use the encoded data stream to feed the data into Apache Kafka for control processing. By controlling the ordering and error detection of the data stream in the message queue, the reliability of message processing is optimized to obtain the queue data stream.
[0350] S3: Performs data control analysis on the queue data stream in the Spark Streaming environment, continuously monitors voltage, current, temperature, and capacity, analyzes time series changes of parameters, controls and identifies abnormal patterns, and generates state time series analysis results;
[0351] S4: Based on the state time series analysis results, control the calculation of demand forecast, define and identify energy demand peaks and sudden power consumption events, control the real-time response processing to events, adjust the output configuration of energy storage units, and obtain event response decisions;
[0352] S5: Use event response decisions to control the energy supply and output of energy storage, adjust the charging and discharging strategies according to demand, monitor and control the energy flow and consumption of the network, analyze and adjust the energy allocation strategy and optimize the energy storage control configuration performance, and establish smart grid energy storage charging and discharging scheduling data results.
[0353] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. A grid-type energy storage system, characterized in that: The system comprises: The data interface and transfer module formats the key parameters of the energy storage unit, such as voltage, current, temperature, and remaining capacity, in JSON based on the energy storage unit data. The data is sent to the central control configuration via the TCP / IP protocol to obtain the encoded data stream, which is then input into the Apache Kafka message queue to obtain the queue data stream. The state analysis and prediction module inputs the queue data stream into a real-time data processing engine based on Spark Streaming, analyzes the voltage, current, temperature and capacity data, obtains state time series analysis results, and dynamically estimates the energy demand of each unit based on the state time series analysis results to obtain demand prediction results; The steps for obtaining the state time series analysis results are specifically as follows: The queue data stream is processed in real time by the Spark Streaming engine, and the transformation function is applied and the processing speed and responsiveness of the data stream are adjusted in combination with the adjustment parameters, using the formula: , Generate processed data streams, is the baseline adjustment factor, and k are dynamic adjustment parameters, Calculate the queue length, represents the processed data stream, Represents the input queue data flow, Conversion functions for real-time processing; Extract key electrical features from the processed data stream, including voltage, current, temperature, and capacity, and weight them based on feature weights, using the formula: , Generate a data stream feature set, where h is the feature extraction function, is the feature weight, Z is the normalization constant, represents the data stream feature set, They are voltage, current, temperature and capacity respectively; The data stream feature set is subjected to time series analysis, using the analysis function combined with the time factor, and adopting the formula: , Generate state time series analysis results, where t represents the time variable, is the decay constant, is the time influence intensity factor, represents the state time series analysis results, represents the data stream feature set, and m represents the time series analysis function; The steps for obtaining the demand forecast result are specifically as follows: The energy data of each unit is integrated from the state time series analysis results, and the historical data is combined with the current data using the enhanced integration function, through the formula: , Generate integrated data, where represents the historical energy demand of the ith unit, n is the total number of units, represents the state time series analysis result, f represents the enhanced integration function, Represents the integrated data; Apply the forecasting model to the integrated data to dynamically calculate the demand in the future time period, through the formula: , Generate forecasted energy demand, where is the time decay constant, t is the time interval from data integration to prediction time, represents the predicted energy demand, represents the integrated data, and p represents the demand forecasting model; The predicted energy demand is formatted and output and matched to the energy unit management configuration through the formula: , Generate demand forecast results, where format is a formatting function. is the formatting parameter, represents the demand forecast result, represents the predicted energy demand; The event response and strategy adjustment module identifies energy demand peaks and sudden batch power consumption events based on the demand forecast results, matches event patterns, uses the Esper engine to perform event processing, obtains event response decisions, uses the event response decisions to adjust energy supply, and obtains an optimized energy allocation plan; The steps for obtaining the event response decision are specifically as follows: Using the demand forecast results, energy demand peaks are defined and sudden batch power consumption events are identified. A threshold identification method is applied, using the formula: , Generate an event pattern where is the dynamically adjusted demand peak threshold, n represents the number of data points in the time series, Represents an event mode, represents the demand forecast result, and x represents the energy demand data at a single time point; The event pattern is applied to the Esper engine for pattern matching to identify key energy events, using the formula: , Generates matching events, where is the importance weight of each event, is the sum of event weights, Indicates the event after matching, Represents the event pattern, match represents the pattern matching function of the Esper engine; Based on the matched events, the decision logic is executed and the identified events are responded to to generate event response decisions, using the formula: , Where k is the adjustment factor, m represents the number of matching events, represents an incident response decision, Represents the event after matching, and decision represents the decision function applied to the matching event; The steps for obtaining the optimized energy allocation scheme are specifically as follows: Extract data from the incident response decision, analyze and identify the energy supply strategy that needs to be adjusted, through the formula: , Generate analyzed energy data, where represents the data of the i-th event response decision, is the adjustment factor, represents the analyzed energy data; According to the analyzed energy data, the energy supply strategy is adjusted to match the current and predicted energy demand, through the formula: , Generate adjusted supply data, where m is the function that adjusts energy supply, represents the adjusted supply data, Indicates the maximum value in the analyzed data. represents the analyzed energy data; To convert the adjusted supply data into an operational strategy, the formula is: , Form an optimized energy allocation plan, where: is the function that converts supply adjustments into policies, It represents the optimization of energy distribution scheme. represents the adjusted supply data, represents a logarithmic transformation for enhancing small value adjustments, represents the logarithm base; The energy dispatching and management module receives the optimized energy allocation plan, adjusts the charging and discharging strategy according to demand, records the energy flow status, analyzes the energy storage charging and discharging efficiency and adaptability, and obtains the intelligent grid energy storage charging and discharging dispatching data results.
2. The grid-type energy storage system according to claim 1, characterized in that: The steps of obtaining the queue data flow are specifically as follows: Based on the energy storage unit data, each key parameter value of the energy storage unit is converted into the attributes of the JSON object through the formatting function, using the formula: , Generate a formatted data packet, where K represents the key parameter. represents the real-time value of parameter k, f represents the data formatting function, represents the weight of parameter k, N represents the total number of parameters, Indicates the formatted data packet; The formatted data packet is sent through the TCP / IP protocol using the sending function, using the formula: , Generates an encoded data stream where represents the i-th formatted data packet, n represents the number of data packets, s represents the sending function, Represents the encoded data stream; The encoded data stream is input into the Apache Kafka message queue, using the queue function, using the formula: , Generate a queue data stream, where represents the encoded data stream, q represents the queue function, L represents the load factor, Represents queue data flow.
3. The grid-type energy storage system according to claim 1, characterized in that: The steps for obtaining the smart grid energy storage charging and discharging scheduling data results are specifically as follows: The optimized energy allocation plan is received and combined with the real-time energy consumption data to form a summary data view, using the formula: , in, Represents aggregated data, represents the i-th strategy result data point, is the corresponding energy consumption data, n is the number of data points; The aggregated data view is used to monitor energy flow, match supply with demand, and adjust energy charging and discharging, using the formula: , Generate energy flow states, where is the weight factor for each data point, v is the global parameter for adjusting the nonlinear response, Indicates the energy flow state, Represents aggregated data; Based on the energy flow state, the efficiency and adaptability of energy distribution are analyzed through the formula: , Generate smart grid energy storage charging and discharging scheduling data results, including: is the efficiency benchmark for each state, are the coefficients used for adaptability analysis, It represents the intelligent grid energy storage charging and discharging scheduling data results, Indicates the energy flow state.
4. A grid-type energy storage control method, characterized in that: The grid-connected energy storage system according to any one of claims 1 to 3 comprises the following steps: Collect the voltage, current, temperature, and remaining capacity data of the energy storage unit through the control interface, format the data into JSON, and control the data transmission process through the TCP / IP protocol, optimize the data packet size and transmission frequency, match the network conditions, and generate a coded data stream; Using the encoded data stream, the data is fed into Apache Kafka for control processing, and the reliability of message processing is optimized by controlling the ordering and error detection of the data stream in the message queue to obtain a queue data stream; Perform data control analysis on the queue data stream in the Spark Streaming environment, continuously monitor voltage, current, temperature and capacity, analyze time series changes of parameters, control and identify abnormal patterns, and generate state time series analysis results; Based on the state time series analysis results, the calculation of the control demand forecast is controlled, the energy demand peak and sudden power consumption events are defined and identified, the real-time response processing to the events is controlled, the output configuration of the energy storage unit is adjusted, and the event response decision is obtained; The event response decision is used to control the energy supply and output of energy storage, adjust the charging and discharging strategy according to demand, control the energy flow and consumption of the network, analyze and adjust the energy allocation strategy and optimize the energy storage control configuration performance, and establish the intelligent grid energy storage charging and discharging scheduling data results.
