A new energy centralized control center SCADA system upgrading and optimizing system

Abstract

This invention discloses an upgrade and optimization system for a new energy centralized control center SCADA system, relating to the field of new energy centralized control center technology. It includes a multi-source heterogeneous fusion acquisition module, a fusion calibration data processing module, a power prediction and linkage safety scheduling module, a multi-source linkage fault analysis module, an operation and maintenance integration and adaptation module, an adaptive extension interface module, and a collaborative management and control module. This ensures the accuracy and consistency of multi-source heterogeneous data acquisition. The power prediction and linkage safety scheduling module effectively balances the real-time performance and security of new energy scheduling through dynamic matching of power fluctuation trends and verification logic. The multi-source linkage fault analysis module, combining communication characteristics with a correlation model of historical fault data, achieves accurate location of linkage faults. Combined with the full lifecycle operation and maintenance logic of the operation and maintenance integration and adaptation module, it significantly improves the targeting and efficiency of equipment operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of new energy centralized control center technology, specifically a new energy centralized control center SCADA system upgrade and optimization system. Background Technology

[0002] The SCADA system of the new energy centralized control center is the core of the cluster monitoring, scheduling and operation and maintenance of new energy power plants. It needs to be adapted to the industry characteristics of random fluctuations in new energy power, diverse types of heterogeneous equipment in power plants and 24-hour uninterrupted operation of the system.

[0003] Existing mature SCADA systems still face several key compatibility bottlenecks when addressing the refined operational needs of centralized control in new energy systems: First, the adaptation and data calibration of multi-source heterogeneous devices lack dynamic coordination capabilities. The adaptation models and data calibration benchmarks for device access are statically configured and cannot be iterated in real time based on device operational stability, making it difficult to guarantee the consistency of multi-source data and thus affecting the accuracy of power prediction and fault diagnosis. Second, the linkage mechanism between power prediction and safe scheduling is imperfect. The safety protection verification process is designed with fixed levels and cannot be dynamically adjusted in conjunction with power fluctuation trends and scheduling command transmission characteristics, making it difficult to achieve a precise balance between scheduling real-time performance and operational safety. Third, the coordination between system expansion and full-process control is insufficient. When adding new heterogeneous devices, the adjustment and calibration of parsing logic and the updating of anomaly identification models are independent of each other, and the security encryption of data interaction between modules and the switching of abnormal links do not form a closed loop, affecting the long-term stable operation of the system after expansion. Summary of the Invention

[0004] The purpose of this invention is to provide an upgrade and optimization solution for a new energy centralized control center SCADA system that can solve problems such as multi-source heterogeneous equipment adaptation, data calibration, power prediction and scheduling linkage, accurate fault diagnosis, operation and maintenance standard adaptation, convenient expansion and secure data interaction, so as to improve the operational stability of the SCADA system, the accuracy of data processing and the efficiency of operation and maintenance scheduling, and ensure the safe and efficient grid connection of new energy power.

[0005] A new energy centralized control center SCADA system upgrade and optimization system includes a multi-source heterogeneous fusion acquisition module, a fusion calibration data processing module, a power prediction linkage safety scheduling module, a multi-source linkage fault analysis module, an operation and maintenance fusion adaptation module, an adaptive expansion interface module, and a collaborative management and control module. Each module achieves collaborative linkage through a preset data interaction link, and the data interaction between modules is fully controlled by the collaborative management module; The multi-source heterogeneous fusion acquisition module is used to extract the core communication features of heterogeneous devices, generate a unified adaptation model based on the core communication features, identify the communication fields of heterogeneous devices based on the unified adaptation model, and simultaneously complete the unified acquisition and parsing of multi-source heterogeneous device data. The acquired and parsed data is then transmitted to the fusion calibration data processing module. The fusion calibration data processing module is used to receive the parsed data transmitted by the multi-source heterogeneous fusion acquisition module, integrate the equipment parameters it collects to generate a unified fusion calibration benchmark, complete the equipment parameter calibration and structured conversion of heterogeneous data through the unified fusion calibration benchmark, simultaneously capture multi-source equipment linkage anomalies, dynamically allocate storage nodes based on data volume and storage node load to realize distributed storage and targeted access of data, and transmit the targeted access data to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module respectively. The power prediction and linkage safety scheduling module is used to collect grid-connected power and environmental parameters, construct a short-term prediction model based on the grid-connected power and environmental parameters, generate power fluctuation trend data through the short-term prediction model, extract scheduling command parameters and classify command levels, match the safety protection verification process corresponding to the command level, and link the scheduling command transmission delay, safety protection verification time and power fluctuation trend data to complete the adjustment of the safety protection verification logic. The multi-source linkage fault analysis module is used to collect historical data of on-site operation and maintenance faults, integrate the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module, construct an equipment parameter anomaly and fault correlation model, complete the comparison of abnormal parameters through the equipment parameter anomaly and fault correlation model, locate the source of multi-source equipment linkage faults and related affected equipment, generate fault handling logic and transmit it to the operation and maintenance fusion adaptation module. The operation and maintenance integration and adaptation module is used to collect equipment operation and maintenance related parameters, combine the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module with its own calculated operation reliability data, generate a unified operation and maintenance integration specification, calculate operation and maintenance priorities based on the equipment operation and maintenance related parameters, equipment communication characteristics and operation reliability data, construct the equipment full life cycle operation and maintenance logic, and synchronize operation and maintenance data to the fusion calibration data processing module. The operation and maintenance data is used for the optimization of the unified fusion calibration benchmark. The adaptive extension interface module is used to extract the communication feature vector of the newly added heterogeneous device, compare the communication feature vector with the unified adaptation model generated by the multi-source heterogeneous fusion acquisition module, adjust the data parsing logic based on the comparison result, complete the access and data integration of the new device, and synchronously update the unified fusion calibration benchmark and parameter linkage anomaly identification model generated by the fusion calibration data processing module. The collaborative management module is used to encrypt the data interaction between modules, build a data interaction anomaly monitoring model, monitor the data interaction status of each module in real time, switch to the backup data interaction link when a data interaction anomaly occurs, clarify the data transmission path of each module, and ensure the collaborative operation of each module.

[0006] The technical solutions provided by the embodiments of this disclosure bring at least the following beneficial effects: It ensures the accuracy and consistency of multi-source heterogeneous data collection; the power prediction and linkage safety scheduling module effectively balances the real-time performance and security of new energy scheduling through dynamic matching of power fluctuation trends and verification logic; the multi-source linkage fault judgment module combines communication characteristics with the correlation model of historical fault data to achieve accurate location of linkage faults, and with the full life cycle operation and maintenance logic of the operation and maintenance integration and adaptation module, it significantly improves the pertinence and efficiency of equipment operation and maintenance.

[0007] Through closed-loop management of encryption processing and abnormal link switching, the security and continuity of data interaction between modules are ensured, meeting the high reliability requirements of 24-hour uninterrupted operation of the new energy centralized control system, and improving the operating accuracy, scheduling efficiency and stability of the SCADA system in the new energy centralized control scenario. Attached Figure Description

[0008] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0009] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0010] Please see Figure 1 This application provides an upgrade and optimization system for a new energy centralized control center SCADA system, including a multi-source heterogeneous fusion acquisition module, a fusion calibration data processing module, a power prediction linkage safety scheduling module, a multi-source linkage fault analysis module, an operation and maintenance fusion adaptation module, an adaptive expansion interface module, and a collaborative management and control module; Each module achieves collaborative linkage through a preset data interaction link, and the data interaction between modules is fully controlled by the collaborative management module; The multi-source heterogeneous fusion acquisition module is used to extract the core communication features of heterogeneous devices, generate a unified adaptation model based on the core communication features, identify the communication fields of heterogeneous devices based on the unified adaptation model, and simultaneously complete the unified acquisition and parsing of multi-source heterogeneous device data. The acquired and parsed data is then transmitted to the fusion calibration data processing module. The fusion calibration data processing module is used to receive the parsed data transmitted by the multi-source heterogeneous fusion acquisition module, integrate the equipment parameters it collects to generate a unified fusion calibration benchmark, complete the equipment parameter calibration and structured conversion of heterogeneous data through the unified fusion calibration benchmark, simultaneously capture multi-source equipment linkage anomalies, dynamically allocate storage nodes based on data volume and storage node load to realize distributed storage and targeted access of data, and transmit the targeted access data to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module respectively. The power prediction and linkage safety scheduling module is used to collect grid-connected power and environmental parameters, construct a short-term prediction model based on the grid-connected power and environmental parameters, generate power fluctuation trend data through the short-term prediction model, extract scheduling command parameters and classify command levels, match the safety protection verification process corresponding to the command level, and link the scheduling command transmission delay, safety protection verification time and power fluctuation trend data to complete the adjustment of the safety protection verification logic. The multi-source linkage fault analysis module is used to collect historical data of on-site operation and maintenance faults, integrate the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module, construct an equipment parameter anomaly and fault correlation model, complete the comparison of abnormal parameters through the equipment parameter anomaly and fault correlation model, locate the source of multi-source equipment linkage faults and related affected equipment, generate fault handling logic and transmit it to the operation and maintenance fusion adaptation module. The operation and maintenance integration and adaptation module is used to collect equipment operation and maintenance related parameters, combine the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module with its own calculated operation reliability data, generate a unified operation and maintenance integration specification, calculate operation and maintenance priorities based on the equipment operation and maintenance related parameters, equipment communication characteristics and operation reliability data, construct the equipment full life cycle operation and maintenance logic, and synchronize operation and maintenance data to the fusion calibration data processing module. The operation and maintenance data is used for the optimization of the unified fusion calibration benchmark. The adaptive extension interface module is used to extract the communication feature vector of the newly added heterogeneous device, compare the communication feature vector with the unified adaptation model generated by the multi-source heterogeneous fusion acquisition module, adjust the data parsing logic based on the comparison result, complete the access and data integration of the new device, and synchronously update the unified fusion calibration benchmark and parameter linkage anomaly identification model generated by the fusion calibration data processing module. The collaborative management module is used to encrypt the data interaction between modules, build a data interaction anomaly monitoring model, monitor the data interaction status of each module in real time, switch to the backup data interaction link when a data interaction anomaly occurs, clarify the data transmission path of each module, and ensure the collaborative operation of each module.

[0011] As an optional embodiment, the communication core features extracted by the multi-source heterogeneous fusion acquisition module are taken from the core fields of the communication frame structure and the data transmission control fields of the heterogeneous devices. After extraction, the communication core features are deduplicated using a feature deduplication algorithm, which is a redundant feature removal algorithm based on Euclidean distance. The specific calculation process is as follows: Calculate the Euclidean distance d = for any two eigenvectors. , where xi and yi are the corresponding components of two feature vectors respectively. When d is less than the feature redundancy threshold, one of the feature vectors is removed and the unique feature vector is retained to finally obtain the communication core feature vector Fi of a single device; fn in Fi is the nth core feature parameter, which is specified by the device communication protocol manual. A weighted fusion algorithm is used to generate a unified adaptation model. The specific calculation steps are as follows: First, collect stability data of the device after it is connected to the system and run continuously. Calculate the compatibility coefficient λi between the device and the system using the stability data, where λi = fault-free running time / total running time. Second, calculate the feature weight Wi = λi / Σλi for each device, where Σλi is the sum of the compatibility coefficients of all devices. Third, obtain the unified adaptation model M = Σ(Wi × Fi) through weighted summation. Calculate the similarity between the communication feature vector of the device to be identified and the unified adaptation model M. When the similarity S reaches the adaptation threshold, the identification of the device communication fields and custom extended fields is completed. The adaptation threshold is obtained by calibration using multi-source device adaptation historical data. The collected data is compared with the real-time operating parameters of the device, which are collected by the device's built-in sensors and transmitted to the multi-source heterogeneous fusion acquisition module. After the comparison is consistent, the unified acquisition and analysis of multi-source heterogeneous device data is completed, and the analyzed data is transmitted to the fusion calibration data processing module.

[0012] As an optional embodiment, the fusion calibration data processing module receives the parsed data X parsed from the multi-source heterogeneous fusion acquisition module, and simultaneously acquires the equipment's factory standard parameter X standard and the field operating parameter X field; X standard is provided by the equipment's factory test report and is directly read by the fusion calibration data processing module; X field is acquired by the field monitoring terminal and transmitted to the fusion calibration data processing module; A weighted fusion algorithm is used to generate a unified fusion calibration benchmark. The specific calculation steps are as follows: First, the initial value of the weight coefficient α is obtained by calibrating the equipment using the factory standard parameters and on-site trial operation data. Second, α is adjusted in real time using the equipment's operational stability data. The adjustment formula is α new = α old × (1 + R stability × adjustment factor), where R stability is the equipment's operational stability coefficient, R stability = error-free running time / total running time, and the adjustment factor is obtained from the calibration benchmark stability test data. Third, the unified fusion calibration benchmark X benchmark is obtained by weighted summation: X benchmark = α × X standard + (1-α) × X on-site. The deviation ΔX between the analytical data X analytical and the unified fusion calibration benchmark X benchmark is calculated as X analytical - X benchmark. The equipment parameter calibration is completed through a deviation correction algorithm. The specific calibration steps are as follows: First, calculate the linear fitting coefficient between ΔX and the equipment operating error, and use the linear fitting coefficient as the correction coefficient K; Second, calculate the calibrated data using the formula Xcalibration = Xanalysis - K × ΔX, and simultaneously complete the standardized calibration of the equipment charging and discharging parameters and output parameters, as well as the structured conversion of heterogeneous data; The equipment charging and discharging parameters and output parameters are uploaded by the equipment controller to the fusion calibration data processing module. The equipment parameters are collected over a continuous period of time. These parameters are collected periodically by the on-site monitoring terminal and transmitted to the fusion calibration data processing module. The time series change rate V = ΔX / Δt is calculated, where Δt is a fixed time interval, which is specified by the system sampling frequency. A parameter linkage anomaly identification model is generated. The specific process is as follows: statistical analysis is performed on the time-series change rate data collected in a continuous period. After removing abnormal time-series change rate data, the extreme value of the remaining data is taken as the normal change range of the parameter linkage anomaly identification model. When the time-series change rate V exceeds the normal change range, the multi-source device linkage anomaly is captured. Based on the cross-feedback between data volume and storage node load, the storage node allocation weight is calculated. The specific calculation steps are as follows: First, divide the data into individual data block sizes Si according to the data storage specifications, and calculate the total data volume ΣSi; Second, calculate the storage node allocation weight W_storage = Si / ΣSi; The storage node load is obtained through cross-monitoring of data transmission rate V_transmission and storage occupancy rate O_occupancy. V_transmission is collected by the link monitoring module and transmitted to the fusion calibration data processing module, while O_occupancy is monitored by the storage node itself and transmitted to the fusion calibration data processing module; The load coefficient L is calculated as ω1 × V_transmission / preset rate + ω2 × O_occupancy, where ω1 and ω2 are load weight coefficients, and ω1 + ω2 = 1. The preset rate is specified by the system communication protocol. When the load factor L reaches the load threshold, new storage nodes are automatically added. Storage nodes are dynamically allocated according to the storage node allocation weight W, realizing distributed storage and targeted access of data. The targeted access data is then transmitted to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module respectively.

[0013] As an optional embodiment, the power prediction and safety scheduling module collects real-time power data P of new energy grid-connected power and environmental parameters. P is collected in real time by the grid-connected monitoring device and transmitted to this module; environmental parameters include light intensity G and wind speed V_wind. Light intensity G is collected by the on-site light sensor and transmitted to this module, and wind speed V_wind is collected by the on-site wind speed sensor and transmitted to this module. The specific construction steps for constructing a short-term prediction model for grid-connected power are as follows: First, collect historical data on grid-connected power and environmental parameters; Second, train the historical data using a linear regression algorithm to obtain model coefficients a, b, and c; Third, construct a short-term prediction model based on the model coefficients a, b, and c: Ppredicted = a × Preal-time + b × G + c × Vwind. The specific calculation steps for generating grid-connected power fluctuation trend data in future time periods are as follows: First, calculate the grid-connected power prediction P(t+1) and prediction P(t) for two consecutive future time periods using a short-term prediction model; Second, calculate the difference between prediction P(t+1) and prediction P(t), which is the grid-connected power fluctuation trend data ΔP prediction = prediction P(t+1) - prediction P(t). The future time period is specified by the grid dispatch requirements. The prediction error is corrected by real-time grid-connected power data feedback. The specific correction steps are as follows: First, calculate the difference ΔP between the real-time grid-connected power P_real-time and the predicted power P_predicted = P_real-time - P_predicted; Second, substitute ΔP into the short-term prediction model and adjust the model coefficients a, b, and c to complete the prediction error correction; Extract the execution time limit T of the dispatching instruction and the equipment coverage area N. T is specified by the power grid dispatching instruction and transmitted to this module, and N is statistically recorded by the system equipment ledger and transmitted to this module. The specific calculation steps for the priority coefficient of the dispatching instruction are as follows: First, set weight coefficients k1 and k2, and k1+k2=1. The weight coefficients k1 and k2 are specified by the power grid dispatching safety level. Second, calculate the priority coefficient of the dispatching instruction using the formula P priority=k1 / T+k2×N. The scheduling instructions are classified into levels according to their priority coefficients, and a security protection verification process corresponding to each level is matched. The core verification nodes and non-core verification nodes are specified by the system security protection specifications. The scheduling instruction transmission delay t_delay and the security protection verification time t_time are calculated in real time. The delay t_delay is collected by the link timing module and transmitted to the module. The time t_time is calculated by the module's own verification unit. The delay impact coefficient K_impact is calculated as t_delay / t_time, and the grid-connected power fluctuation threshold ΔP_threshold is calculated as k × P_real-time, where k is the fluctuation coefficient, which is specified by the grid stability requirements. When the delay impact coefficient K reaches the delay impact threshold, and the grid-connected power fluctuation trend data ΔP prediction exceeds the grid-connected power fluctuation threshold ΔP, the security protection verification logic of the corresponding instruction is adjusted. The specific adjustment process is as follows: switch the high-level verification process to the low-level verification process, remove the corresponding number of non-core verification nodes according to the system security protection specifications, and complete the adjustment of the security protection verification logic.

[0014] As an optional embodiment, the multi-source linkage fault judgment module collects historical data of on-site operation and maintenance faults (fault D) and integrates the device communication characteristics (communication F) transmitted by the multi-source heterogeneous fusion acquisition module; fault D is provided by the historical records of the operation and maintenance management system and transmitted to this module, while communication F is extracted by the multi-source heterogeneous fusion acquisition module and directly transmitted to this module. The following steps are taken to construct a correlation model between equipment parameter anomalies and faults: First, data correlation is performed between fault D and communication F; second, the correlation data is trained using a machine learning algorithm to obtain model coefficients mi and ij; third, based on the model coefficients mi and ij, a correlation model between equipment parameter anomalies and faults is constructed, with the model expression being Y_fault = Σ(mi × X_anomaly + ij × F_communication); current equipment anomaly parameter X_current_anomaly is collected, and X_current_anomaly is filtered by the fusion calibration data processing module and transmitted to this module; the matching degree between X_current_anomaly and the anomaly parameters in the correlation model between equipment parameter anomalies and faults is calculated. When the matching degree M reaches the matching threshold, the source of the multi-source device linkage fault and the associated affected devices are located. The specific location process is as follows: based on the correspondence between abnormal parameters and fault types in the device parameter anomaly and fault association model, combined with the calculation result of the matching degree M, the fault source device and the associated affected devices that have data interaction with the fault source device are determined. Based on the fault source parameters, calculate the related equipment parameter adjustment value ΔX adjustment = X standard - X current anomaly, where X standard is provided by the equipment factory inspection report and transmitted to this module; Generate fault handling logic that includes fault diagnosis path and related equipment parameter adjustment logic, and transmit the fault handling logic to the operation and maintenance integration and adaptation module.

[0015] As an optional embodiment, the operation and maintenance integration and adaptation module collects the equipment operation and maintenance cycle T and the fault handling standard parameter X. T-maintenance is defined in the maintenance manual provided by the equipment manufacturer and transmitted to this module; X-handling is defined in the maintenance management specifications and transmitted to this module. The device communication characteristics F-communication are received from the multi-source heterogeneous fusion acquisition module, and the device operation reliability data R-reliability is calculated. R-reliability = fault-free running time / total running time. The fault-free running time and total running time are counted by the timing unit of the module itself. The unified operation and maintenance integration specification is generated through the following steps: First, a weight coefficient β is set, which is determined by the requirements of operation and maintenance efficiency and equipment reliability; Second, the specification parameters are calculated using the formula X_specification = β × X_handling + (1-β) × F_communication-related parameters; Third, the unified operation and maintenance integration specification is generated based on the specification parameters. The calculation steps for the equipment failure rate Ffailure and the operational impact Iimpact are as follows: First, count the number of failures and the total runtime. The number of failures is counted by the operation and maintenance management system and transmitted to this module, and the total runtime is counted by the timing unit of this module itself. Second, calculate the failure rate using the formula Ffailure = number of failures / total runtime. The calculation steps for the impact of I are as follows: First, collect the output ratio of the equipment and the number of equipment. The output ratio of the equipment is collected by the grid-connected monitoring device and transmitted to this module, and the number of equipment is counted by the system equipment ledger and transmitted to this module; Second, calculate the degree of operational impact using the formula I impact = output ratio of equipment × number of equipment. The specific steps for generating the operation and maintenance priority linkage calculation model are as follows: First, set weight coefficients k3 and k4, and k3+k4=1. The weight coefficients k3 and k4 are determined by the importance level of the equipment. Second, construct the operation and maintenance priority linkage calculation model using the formula O priority = k3×F fault + k4×I impact. The operation and maintenance priority is calculated through the operation and maintenance priority linkage calculation model, the operation and maintenance logic of the entire life cycle of the equipment is constructed, and the operation and maintenance data is synchronized to the fusion calibration data processing module. The operation and maintenance data is used to adjust the weight coefficient α of the unified fusion calibration benchmark.

[0016] As an optional embodiment, the adaptive extension interface module adopts a standardized interface architecture, extracts the communication feature vector Fnew of the newly added heterogeneous device, and Fnew is output by the communication interface of the newly added device and transmitted to the module. The similarity between F and the unified adaptation model M generated by the multi-source heterogeneous fusion acquisition module is calculated. The similarity calculation steps are the same as those described above, and the similarity S is obtained. When the similarity S reaches the matching similarity threshold, the data parsing logic is adjusted. The specific adjustment steps are as follows: First, set the matching coefficient γ, which is obtained from the matching test data of the new device and the system; Second, extract the existing parsing parameters currently used by the multi-source heterogeneous fusion acquisition module; Third, calculate the adjusted parsing logic parameters using the formula X_parsing = γ × F_new + (1-γ) × existing parsing parameters; Fourth, complete the adjustment of the data parsing logic based on the adjusted parsing logic parameters, thereby completing the access of the new device and data integration. The newly added device parameter X is collected synchronously. X is collected by the sensor of the newly added device itself and transmitted to this module. The deviation ΔX = X - X reference is calculated between X and the unified fusion calibration reference X reference. X reference is transmitted to this module by the fusion calibration data processing module. The unified fusion calibration benchmark is adjusted in the following steps: First, an adjustment coefficient is set, which is obtained from the calibration benchmark stability test data; Second, the adjusted unified fusion calibration benchmark is calculated using the formula Xbenchmark new = Xbenchmark + ΔXnew × adjustment coefficient; Third, the adjusted unified fusion calibration benchmark is transmitted to the fusion calibration data processing module. The normal variation range of the parameter linkage anomaly identification model is updated synchronously, and the updated normal variation range is transmitted to the fusion calibration data processing module.

[0017] As an optional implementation, the collaborative management module uses a symmetric encryption algorithm to encrypt the data interaction between modules; The specific encryption steps are as follows: First, generate an encryption key K, which is generated by the system key management module and updated periodically, and transmit K to the collaborative management module; Second, obtain the plaintext data D exchanged between the modules; Third, encrypt D using the symmetric encryption algorithm E, with the encryption formula C=E(K,D), where C is the encrypted ciphertext, which is used for data transmission between the modules. The data interaction anomaly linkage monitoring model is constructed in the following steps: First, a rate threshold is set, which is specified by the system communication protocol; Second, the data interaction anomaly linkage monitoring model is constructed based on the data transmission rate and the check code; The data transmission rate V_transmission and the check code CRC are calculated in real time. V_transmission is collected by the link monitoring module and transmitted to the module. The data byte datai is extracted from the data transmission frame. When the data transmission rate V is lower than the rate threshold, or the CRC check does not match, it is determined that the data interaction is abnormal. When a data interaction abnormality occurs, it switches to the backup data interaction link. The backup data interaction link is designed and deployed by the system redundancy, and the switching process is fed back to the collaborative management and control module in real time. The data transmission paths for each module are defined as follows: multi-source heterogeneous fusion acquisition module, fusion calibration data processing module, multi-source linkage fault analysis module / power prediction linkage safety scheduling module, and operation and maintenance fusion adaptation module / adaptive extension interface module. Link monitoring ensures the coordinated operation of all modules. The link monitoring is performed by the link monitoring module, and the monitoring data is transmitted to the collaborative management and control module in real time.

[0018] As an optional embodiment, the Vmin and Vmax involved in the normal variation range of the parameter linkage anomaly identification model are statistically derived from the time-series change rate calculated by the device parameters collected at a fixed frequency within a continuous time period, and the fixed collection frequency is specified by the system sampling specification. The specific statistical steps are as follows: First, collect equipment parameters within a continuous time period at a fixed collection frequency; second, calculate the time series change rate at each collection moment using the formula V=ΔX / Δt; third, remove outliers from all time series change rate data using the 3σ criterion; fourth, after removing outliers, take the minimum value of the remaining data as Vmin and the maximum value as Vmax to form the normal change range of the parameter linkage anomaly identification model.

[0019] As an optional embodiment, the model coefficients a, b, and c of the short-term grid-connected power prediction model are obtained by training the model using a linear regression algorithm based on historical data of grid-connected power, solar irradiance, and wind speed. The specific training steps are as follows: First, collect historical data on grid-connected power, solar irradiance, and wind speed at a fixed collection frequency, as specified in environmental monitoring standards. Second, divide the historical data into a training set and a validation set. The training set is used for model training, and the validation set is used for model validation. Third, set the number of iterations and the convergence threshold for the linear regression algorithm, as specified by the model training requirements. Fourth, substitute the training set data into the linear regression equation and iteratively optimize the model coefficients a, b, and c using the gradient descent method until the number of iterations reaches the set value or the iteration error is less than the convergence threshold, thus completing the training of the model coefficients. Fifth, verify the accuracy of the model coefficients using the validation set data. After successful verification, determine the model coefficients a, b, and c, which are used to construct a short-term prediction model for grid-connected power.

[0020] Working principle Multi-source heterogeneous data acquisition and analysis: The multi-source heterogeneous fusion acquisition module first extracts the core fields of the communication frame structure and data transmission control fields of each heterogeneous device as core communication features. Redundant features are removed by a redundant feature elimination algorithm based on Euclidean distance to obtain the core feature vector of a single device's communication. Then, a weighted fusion algorithm is used to calculate feature weights by combining the compatibility coefficient between the device and the system, generating a unified adaptation model. Based on this unified adaptation model, the communication field identification of the heterogeneous device to be identified is completed. After comparing the acquired data with the real-time operating parameters of the device, the unified acquisition and analysis of multi-source heterogeneous device data is completed, and the analyzed data is transmitted to the fusion calibration data processing module.

[0021] Data calibration, anomaly detection, and storage: The fusion calibration data processing module receives parsed data transmitted from the multi-source heterogeneous fusion acquisition module, synchronously collects equipment factory standard parameters and field operating parameters, constructs a unified fusion calibration benchmark using a weighted fusion algorithm, and adjusts it in real time; it calculates the deviation between the parsed data and the calibration benchmark, and completes data calibration and heterogeneous data structure conversion through a deviation correction algorithm; it also collects equipment parameters over continuous periods, calculates the time-series change rate, determines the normal change range after statistically eliminating abnormal data, and detects multi-source equipment linkage anomalies; it calculates storage node allocation weights based on data volume and storage node load, dynamically allocates storage nodes, realizes distributed data storage and targeted access, and transmits the data to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module respectively.

[0022] Power prediction and safety dispatch linkage: The power prediction linkage safety dispatch module collects real-time grid-connected power data and environmental parameters such as sunlight and wind speed. Based on historical data, it trains model coefficients using a linear regression algorithm to construct a short-term prediction model. This model calculates grid-connected power fluctuation trend data for future periods and corrects prediction errors by combining it with real-time grid-connected power data. It extracts the execution time limit and equipment coverage of dispatch instructions, calculates the priority coefficient of dispatch instructions and classifies instruction levels, and matches corresponding safety protection verification processes. It calculates the transmission delay and verification time of dispatch instructions in real time, links them with power fluctuation trend data, and adjusts the safety protection verification logic when the delay impact coefficient and power fluctuation exceed the corresponding threshold to ensure dispatch safety and efficiency.

[0023] Multi-source equipment fault assessment and handling: The multi-source linkage fault assessment module collects historical data of on-site operation and maintenance faults, integrates the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module, correlates the two data and trains model coefficients through machine learning algorithms to construct an equipment parameter anomaly-fault correlation model; receives the current equipment anomaly parameters transmitted by the fusion calibration data processing module, calculates the matching degree between them and the anomaly parameters in the correlation model, when the matching degree reaches the threshold, locates the fault source and related affected equipment, calculates the parameter adjustment value of the related equipment, generates fault handling logic and transmits it to the operation and maintenance fusion adaptation module.

[0024] Equipment Operation and Maintenance Integration Adaptation and Optimization: The operation and maintenance integration adaptation module collects operation and maintenance-related parameters such as equipment operation and maintenance cycle and fault handling standard parameters, receives equipment communication characteristics transmitted by the multi-source heterogeneous integration acquisition module, and calculates equipment operation reliability data; based on the above parameters, it generates a unified operation and maintenance integration specification, calculates the equipment failure rate and operational impact, constructs an operation and maintenance priority linkage calculation model, determines operation and maintenance priorities, and constructs the equipment's full life cycle operation and maintenance logic; it synchronizes operation and maintenance data to the integration calibration data processing module for optimizing the unified integration calibration benchmark.

[0025] System adaptive expansion and adaptation: The adaptive expansion interface module adopts a standardized interface architecture, extracts the communication feature vectors of newly added heterogeneous devices, and performs similarity calculation with the unified adaptation model generated by the multi-source heterogeneous fusion acquisition module; when the similarity reaches the threshold, the data parsing logic is adjusted to complete the access of the new device and data integration; the parameters of the new device are collected synchronously, the deviation between them and the unified fusion calibration benchmark is calculated, the calibration benchmark is adjusted and the normal change range of the parameter linkage anomaly identification model is updated to realize the adaptive expansion of the system.

[0026] Collaborative Management and Security Assurance of Each Module: The collaborative management module oversees the data interaction between modules throughout the entire process, employs a symmetric encryption algorithm to encrypt the data, generates and periodically updates encryption keys to ensure data transmission security, and constructs a data interaction anomaly linkage monitoring model to calculate the data transmission rate and checksum in real time, determine data interaction anomalies and switch to a backup data interaction link, clarify the data transmission path of each module, and ensure the coordinated and stable operation of each module through link monitoring, thereby achieving efficient, secure and stable operation of the entire system.

[0027] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A new energy centralized control center SCADA system upgrade and optimization system, characterized in that, It includes a multi-source heterogeneous fusion acquisition module, a fusion calibration data processing module, a power prediction and linkage safety scheduling module, a multi-source linkage fault analysis module, an operation and maintenance fusion adaptation module, an adaptive expansion interface module, and a collaborative management and control module; The multi-source heterogeneous fusion acquisition module is used to extract the core communication features of heterogeneous devices, complete the unified acquisition and parsing of data from multiple heterogeneous devices, and transmit the acquired and parsed data to the fusion calibration data processing module. The fusion calibration data processing module is used to receive the parsed data transmitted by the multi-source heterogeneous fusion acquisition module and transmit it to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module. The power prediction and safety scheduling module is used to collect grid-connected power and environmental parameters, extract scheduling command parameters and classify command levels, match the safety protection verification process corresponding to the command level, and link scheduling command transmission delay, safety protection verification time and power fluctuation trend data. The multi-source linkage fault analysis module is used to collect historical data of on-site operation and maintenance faults, integrate the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module, and construct a model of the correlation between abnormal equipment parameters and faults. The operation and maintenance integration and adaptation module is used to collect equipment operation and maintenance related parameters. It combines the equipment communication characteristics transmitted by the multi-source heterogeneous fusion acquisition module with its own calculated operation reliability data to construct the equipment's full life cycle operation and maintenance logic and synchronize the operation and maintenance data to the fusion calibration data processing module.

2. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, The core communication features extracted by the multi-source heterogeneous fusion acquisition module are taken from the core fields of the communication frame structure and the data transmission control fields of the heterogeneous devices. After extraction, the core communication features are deduplicated using a feature deduplication algorithm. A weighted fusion algorithm is used to generate a unified adaptation model. The collected data is compared with the real-time operating parameters of the device. The real-time operating parameters of the device are collected by the device's built-in sensors and transmitted to the multi-source heterogeneous fusion acquisition module. After the comparison is consistent, the unified acquisition and parsing of multi-source heterogeneous device data is completed, and the parsed data is transmitted to the fusion calibration data processing module.

3. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 2, characterized in that, The fusion calibration data processing module receives the parsed data X parsed from the multi-source heterogeneous fusion acquisition module, and simultaneously acquires the equipment's factory standard parameter X standard and the field operating parameter X field; X standard is provided by the equipment's factory test report and is directly read by the fusion calibration data processing module; The X data is collected by the on-site monitoring terminal and transmitted to the fusion calibration data processing module; A weighted fusion algorithm is used to generate a unified fusion calibration benchmark. Equipment parameter calibration is completed through a deviation correction algorithm. Equipment parameters are collected over a continuous period of time. These equipment parameters are collected periodically by the field monitoring terminal and transmitted to the fusion calibration data processing module. The time series change rate V = ΔX / Δt is calculated, where Δt is a fixed time interval, which is specified by the system sampling frequency. A parameter-linked anomaly detection model is generated. Based on the cross-feedback between data volume and storage node load, the weights for allocating storage nodes are calculated. The specific calculation steps are as follows: The first step is to divide the data into individual data block sizes Si according to the data storage specifications and calculate the total data volume ΣSi. The second step is to calculate the storage node allocation weight W_storage = Si / ΣSi. The storage node load is obtained by cross-monitoring the data transmission rate V_transmission and the storage occupancy rate O_occupancy. V_transmission is collected by the link monitoring module and transmitted to the fusion calibration data processing module, while O_occupancy is monitored by the storage node itself and transmitted to the fusion calibration data processing module. The load factor L is calculated as L = ω1 × V transmission / preset rate + ω2 × 0 occupancy, where ω1 and ω2 are load weighting coefficients, and ω1 + ω2 = 1. The preset rate is specified by the system communication protocol. When the load factor L reaches the load threshold, new storage nodes are automatically added, and storage nodes are dynamically allocated according to the weight W assigned to the storage nodes to realize distributed storage and targeted access of data. The targeted access data is then transmitted to the multi-source linkage fault judgment module and the power prediction linkage safety scheduling module respectively.

4. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, The power prediction and safety dispatch module collects real-time data on the grid-connected power of new energy sources (P) and environmental parameters. P is collected in real time by the grid-connected monitoring device and transmitted to this module; environmental parameters include light intensity G and wind speed Vwind; light intensity G is collected by the on-site light sensor and transmitted to this module; wind speed Vwind is collected by the on-site wind speed sensor and transmitted to this module; A short-term prediction model for grid-connected power is constructed to generate grid-connected power fluctuation trend data in future periods. The prediction error is corrected by real-time grid-connected power data feedback. The execution time limit T of the dispatching instruction and the equipment coverage N are extracted. T is specified by the grid dispatching instruction and transmitted to this module, and N is statistically recorded by the system equipment ledger and transmitted to this module. Calculate the priority coefficient of the scheduling instruction, classify the instruction level according to the priority coefficient, and match the security protection verification process corresponding to the instruction level. The core verification node and non-core verification node are specified by the system security protection specification. The real-time calculation of the scheduling instruction transmission delay t_delay and the security protection verification time t_time is performed; t_delay is collected by the link timing module and transmitted to this module, and t_time is calculated by the module's own verification unit. Calculate the impact of delay coefficient K. When the impact of delay coefficient K reaches the delay impact threshold, and the predicted grid-connected power fluctuation trend data ΔP exceeds the grid-connected power fluctuation threshold ΔP, adjust the security protection verification logic of the corresponding instruction. The specific adjustment process is as follows: switch the high-level verification process to the low-level verification process, remove the corresponding number of non-core verification nodes according to the system security protection specifications, and complete the adjustment of the security protection verification logic.

5. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, The multi-source linkage fault analysis module collects historical data of on-site operation and maintenance faults (fault D) and integrates the device communication characteristics (communication F) transmitted by the multi-source heterogeneous fusion acquisition module. Fault D is provided by the historical records of the operation and maintenance management system and transmitted to this module, while communication F is extracted by the multi-source heterogeneous fusion acquisition module and directly transmitted to this module. Construct a correlation model between abnormal equipment parameters and faults, and calculate the matching degree between the current abnormality of X and the abnormal parameters in the correlation model between abnormal equipment parameters and faults; When the matching degree M reaches the matching threshold, the source of the multi-source device linkage fault and the associated affected devices are located. The specific location process is as follows: based on the correspondence between abnormal parameters and fault types in the device parameter anomaly and fault association model, combined with the calculation result of the matching degree M, the fault source device and the associated affected devices that have data interaction with the fault source device are determined. Based on the fault source parameters, calculate the related equipment parameter adjustment value ΔX adjustment = X standard - X current anomaly, where X standard is provided by the equipment factory inspection report and transmitted to this module; Generate fault handling logic that includes fault diagnosis path and related equipment parameter adjustment logic, and transmit the fault handling logic to the operation and maintenance integration and adaptation module.

6. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, The operation and maintenance integration and adaptation module collects equipment operation and maintenance cycle T, and standard parameters X for fault handling; T-maintenance is defined in the maintenance manual provided by the equipment manufacturer and transmitted to this module; X-handling is defined in the maintenance management specifications and transmitted to this module. The device communication characteristics F-communication are received from the multi-source heterogeneous fusion acquisition module, and the device operation reliability data R-reliability is calculated. R-reliability = fault-free runtime / total runtime; the fault-free runtime and total runtime are statistically analyzed by the timing unit of the module itself. Generate a unified operation and maintenance integration standard, calculate the equipment failure rate Ffailure and the operational impact Iimpact, and the calculation steps for Ffailure are as follows: The calculation steps for the impact of I are as follows: Generate an operation and maintenance priority linkage calculation model; the specific generation steps are as follows: First, set weight coefficients k3 and k4, and k3+k4=1, wherein the weight coefficients k3 and k4 are specified by the importance level of the equipment; Second, construct the operation and maintenance priority linkage calculation model through the formula O priority=k3×F fault+k4×I impact; The operation and maintenance priority is calculated through the operation and maintenance priority linkage calculation model, the operation and maintenance logic of the entire life cycle of the equipment is constructed, and the operation and maintenance data is synchronized to the fusion calibration data processing module. The operation and maintenance data is used to adjust the weight coefficient α of the unified fusion calibration benchmark.

7. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, It also includes an adaptive extension interface module, which is used to extract the communication feature vector of the newly added heterogeneous device and compare the communication feature vector with the unified adaptation model generated by the multi-source heterogeneous fusion acquisition module; The adaptive extension interface module adopts a standardized interface architecture, extracts the communication feature vector F_new from the newly added heterogeneous device, and F_new is output by the communication interface of the newly added device and transmitted to this module. The similarity between F and the unified adaptation model M generated by the multi-source heterogeneous fusion acquisition module is calculated. The similarity calculation steps are the same as those described above, and the similarity S is obtained. When the similarity S reaches the appropriate similarity threshold, adjust the data parsing logic: The newly added device parameter X is collected synchronously. X is collected by the sensor of the newly added device itself and transmitted to this module. The deviation ΔX = X - X reference is calculated between X and the unified fusion calibration reference X reference. X reference is transmitted to this module by the fusion calibration data processing module. Adjust the unified fusion calibration benchmark, synchronously update the normal variation range of the parameter linkage anomaly identification model, and transmit the updated normal variation range to the fusion calibration data processing module.

8. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 1, characterized in that, It also includes a collaborative management and control module, which is used to encrypt the data interaction between modules, build a data interaction anomaly monitoring model, monitor the data interaction status of each module in real time, and switch to a backup data interaction link when a data interaction anomaly occurs. The collaborative management module uses a symmetric encryption algorithm to encrypt data interactions between modules; The data interaction anomaly linkage monitoring model is constructed through the following steps: First, a rate threshold is set, which is specified by the system communication protocol; Second, the data interaction anomaly linkage monitoring model is constructed based on the data transmission rate and the checksum; The data transmission rate Vtransmission and the checksum CRC are calculated in real time. Vtransmission is collected by the link monitoring module and transmitted to the module; The data byte datai is extracted from the data transmission frame. When the data transmission rate V is lower than the rate threshold, or the CRC check does not match, it is determined that the data interaction is abnormal. When a data interaction anomaly occurs, the system switches to a backup data interaction link. The backup data interaction link is designed and deployed with system redundancy, and the switching process is fed back to the collaborative management and control module in real time. The data transmission paths for each module are defined as follows: multi-source heterogeneous fusion acquisition module, fusion calibration data processing module, multi-source linkage fault analysis module / power prediction linkage safety scheduling module, and operation and maintenance fusion adaptation module / adaptive extension interface module. Link monitoring ensures the coordinated operation of all modules. The link monitoring is performed by the link monitoring module, and the monitoring data is transmitted to the collaborative management and control module in real time.

9. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 3, characterized in that, The normal variation range of the parameter linkage anomaly identification model, including Vmin and Vmax, is calculated from the time-series change rate obtained by collecting equipment parameters at a fixed frequency within a continuous time period. The fixed collection frequency is specified by the system sampling specification.

10. The SCADA system upgrade and optimization system for a new energy centralized control center according to claim 4, characterized in that, The model coefficients a, b, and c of the short-term prediction model for grid-connected power are obtained by training the model using a linear regression algorithm based on historical data of grid-connected power, solar irradiance, and wind speed. The specific training steps are as follows: First, collect historical data on grid-connected power, solar irradiance, and wind speed at a fixed collection frequency, which is specified by environmental monitoring standards; Second, divide the historical data into a training set and a validation set, with the training set used for model training and the validation set used for model validation; Third, set the number of iterations and the convergence threshold for the linear regression algorithm, which are specified by the model training requirements; Fourth, substitute the training set data into the linear regression equation and iteratively optimize the model coefficients a, b, and c using the gradient descent method until the number of iterations reaches the set value or the iteration error is less than the convergence threshold, thus completing the training of the model coefficients. The fifth step is to verify the accuracy of the model coefficients using validation set data. Once the verification is successful, the model coefficients a, b, and c are determined and used to construct a short-term prediction model for grid-connected power.

Images