A virtual power plant regulation method suitable for medical systems

Abstract

The present invention relates to the field of energy management technology, specifically a virtual power plant regulation method suitable for medical systems, comprising the following steps: obtaining the start and stop times of energy regions and marking and classifying them, extracting electricity price nodes for comparison and operation status, generating a node correspondence table, extracting load node comparison directions to generate a delay list, judging temperature and cost differences to generate conversion marks, extracting regional data to judge attribution to generate a path table, and identifying operation duration to generate a regulation plan. In the present invention, by combining the direction of load and price changes, screening response deviation nodes, enhancing the strength of anomaly identification, utilizing the cross-judgment of temperature curves and energy cost differences to improve the rationality of path switching, introducing collaborative analysis of cost differences and spatial attributes, refining path attribution adjustments, generating path call information based on continuous activation duration, supplementing periodic tracking capabilities, and overall enhancing the dynamic adaptability and control accuracy of the regulation strategy.

Description

Technical Field

[0001] The present invention relates to the field of energy management technology, and in particular to a virtual power plant regulation method applicable to medical systems. Background Art

[0002] The field of energy management technology encompasses a management technology system for the entire process of energy acquisition, conversion, allocation, use, and monitoring. Its core content is to achieve scheduling optimization and coordinated control of various energy media in the energy system, such as electricity, heat, and gas. This technical field aims to improve the efficiency and safety of energy use by building an energy management platform that integrates intelligent perception, real-time data acquisition, optimized scheduling decisions, and coupled control across energy systems. Against the backdrop of increasing demand for multi-energy complementarity, energy management technology is increasingly developing towards virtual power plants, multi-energy collaborative energy routing, and demand response. In particular, in specific scenarios such as smart buildings, industrial parks, and medical systems, the requirements for complexity and coordination are higher.

[0003] Among them, the virtual power plant regulation method applicable to the medical system refers to a scheduling method for the coordinated demand for cooling, heating and electricity in hospital-type load scenarios. The technical matters covered by this patent subject include obtaining real-time natural gas prices and electricity market price data based on the changing rules of the hospital's cooling and heating loads to calculate the energy efficiency ratio of natural gas in the heating and cooling links and perform difference analysis with electricity prices, and then construct a time series scheduling plan based on the obtained economic difference combined with the load forecast results to determine whether to enable or suspend the operation of some virtual power plant subsystems. Its method generally adopts energy price model prediction time segment load aggregation optimization algorithm and energy conversion path fixed mapping and other means to complete the virtual power plant regulation strategy formulation and control instruction generation.

[0004] Existing technologies often construct dispatching logic based on single-dimensional price and load forecasting models for energy media. This lacks a point-to-point temporal correspondence between energy behavior characteristics and electricity price nodes, resulting in weak dispatching responses in scenarios with rhythmic electricity prices. Load change state identification relies on statistical trend estimation, failing to establish a discriminative path between load change direction and price trends, resulting in insufficient ability to discern the direction of load response. Energy switching logic lacks multivariate cross-judgment support, and a dynamic linkage between temperature feedback and cost structure changes is not established, leading to hysteresis or misjudgment in path switching. The energy attribution path partitioning mechanism focuses on identifying spatial attribute labels and fails to incorporate a combined analysis of energy cost changes and energy call behavior, resulting in low path attribution classification accuracy. A lack of a cycle identification method for the call continuity of energy channels results in fragmented and discrete operational behavior, impacting the strategic coherence and control stability of virtual power plants in continuous regulation tasks. These deficiencies limit the dispatching system's structural perception, behavior judgment, and path decision-making capabilities in complex collaborative scenarios. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a virtual power plant regulation method suitable for medical systems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a virtual power plant adjustment method applicable to a medical system, comprising the following steps:

[0007] S1: Obtain the start and stop times of the energy zones in the medical scenario, classify and mark them by hour, extract the electricity price change nodes, sequentially compare the operating status with the electricity price changes, calculate the point-to-point interval duration, and generate a corresponding relationship table between the operating and electricity price nodes;

[0008] S2: Extracting load change nodes based on the table of correspondence between operation and electricity price nodes, obtaining actual load data for the corresponding period, comparing load and electricity price change directions, identifying consistent and inconsistent records, and generating a response delay list;

[0009] S3: Extract the hot water node in the response delay list, obtain the return temperature sequence and the electricity and gas prices of the same period, determine whether the cost difference and temperature change meet the specified range, and if so, set it as an electricity channel and generate an energy source conversion execution flag;

[0010] S4: Locate the node area based on the energy source conversion execution mark, extract lighting, electric cooling, temperature control data and original energy records, and determine whether to trigger a change based on the cost difference. If triggered, classify it as a power path and generate a power supply level path table;

[0011] S5: Extract the change nodes in the power supply level path table, obtain the channel start and stop time, calculate the continuous operation time, identify the path call information in combination with the node number, and form a virtual power plant adjustment plan.

[0012] As a further solution of the present invention, the operation and electricity price node correspondence table includes start and stop time classification labels, electricity price node time series, comparison time labels of electricity price and operation status, operation interval type labels, and trend change types. The response delay list includes load change categories, electricity price change directions, change direction consistency labels, direction inconsistent node numbers, and inconsistent record archiving labels. The energy source conversion execution mark includes hot water heating node numbers, electricity and natural gas cost difference ranges, hot water temperature fluctuation ranges, and electricity channel switching marks. The power supply level path table includes node energy path classifications, area identification codes, lighting and temperature control category data, electricity cost difference ranges, and electricity path level labels. The virtual power plant adjustment plan includes node number indexes, energy channel operation start and end times, operation cycle duration, and path continuous call labels.

[0013] As a further solution of the present invention, the specific steps of S1 are:

[0014] S101: Obtain the start and stop times of energy usage areas in the medical scenario, divide the start and stop signals by hour, determine the relationship between the start and stop points and the corresponding hourly intervals, mark the operating status of each hour, and generate an hourly start and stop status type classification result;

[0015] S102: Based on the hourly start / stop status type classification result, detect the difference in electricity value at adjacent time points in the electricity price sequence, extract the change node, pair it with the hourly start / stop status, calculate the electricity price change correction value, and obtain the interval between the operating state and the electricity price node;

[0016] S103: Call the interval between the operating status and the electricity price node, compare the start-stop trend direction and the electricity price change direction within the interval, mark the time period status according to the trend consistency classification, and obtain the operating trend and electricity price node matching type table.

[0017] As a further solution of the present invention, the formula for calculating the electricity price change correction value is specifically:

[0018] ;

[0019] in, Represents the electricity price change correction value of the current hour i, represents the electricity price value in the i-th hour, Indicates the electricity price value of the previous hour. represents the operating load value at hour i, represents the weight factor corresponding to the start-stop state in the i-th hour, It represents the average value of the operating load value in the continuous period related to the node. It represents the average value of the electricity price in consecutive time periods related to the node.

[0020] As a further solution of the present invention, the specific steps of S2 are:

[0021] S201: Based on the operation trend and electricity price node matching type table, filter the nodes marked as load increase and decrease, extract the time value of the corresponding time period, obtain the load collection sequence data within the time period, call the node time range and the load collection sequence for interval matching, extract the load change of the node corresponding to the time period, and obtain a load change value list;

[0022] S202: Calling the load change value list, classifying and comparing the load change direction and operation trend of each node with the corresponding electricity price change direction in the electricity price node matching type table, determining whether the directions are consistent, marking the consistent and inconsistent states according to the node number, and generating a load and electricity price trend consistency classification result;

[0023] S203: According to the load and electricity price trend consistency classification results, extract all nodes marked with inconsistent directions, call the node number and load change value, and electricity price change direction to establish a corresponding list structure, sort and archive them by node time, and generate a response delay list.

[0024] As a further solution of the present invention, the specific steps of S3 are:

[0025] S301: Extracting the hot water heating node numbers from the response delay list, obtaining the time period corresponding to each number at the return end, collecting temperature series data at the hot water return end within the time period to which the node belongs, arranging the correspondence between time and temperature by hour, establishing an association structure between the hot water node and the time period temperature, and generating a hot water temperature series distribution result;

[0026] S302: The hot water temperature sequence distribution result is called, and the unit price sequence data of the electricity cost and the natural gas cost in the same time period are collected. The cost difference is calculated according to the time. The calculation result is combined with the temperature value at the time point to determine whether it meets the preset cost difference threshold and the temperature range limit, and the time node number that meets the energy switching determination condition is generated;

[0027] S303: Based on the time node number that meets the energy switching determination condition, set the heating path identification field according to the node number, set the heating path state to the electric channel state, and generate an energy source conversion execution mark.

[0028] As a further solution of the present invention, the specific steps of S4 are:

[0029] S401: Based on the node number in the energy source conversion execution tag, locate the spatial area to which the corresponding node belongs, collect lighting equipment power data, electric cooling load records, and switching time series of temperature control equipment in the area, merge the status records of all devices in the current time period according to the node number, and generate a spatial area device status structure;

[0030] S402: Calling the spatial area device status structure, extracting the energy call record corresponding to the node number, collecting the unit price sequence of electricity and natural gas costs in the current period, calculating the difference between the electricity and natural gas costs, and comparing it with the ownership change cost threshold set for the node number to determine whether the change condition is triggered and generate the ownership change trigger status;

[0031] S403: According to the ownership change trigger status, filter the node numbers that have triggered the change conditions, set the node power status to the power path, and update the ownership status of the level field to the power main path level, and establish a power supply level path table.

[0032] As a further solution of the present invention, the formula for calculating the difference between electricity and natural gas costs is specifically:

[0033] ;

[0034] in, Representative number is The nodes in the period The difference between electricity and natural gas costs within Representative number is The nodes in the period The number of energy-related call devices, Representative number is The node of Devices in the time period The unit price of electricity within Representative number is The node of Devices in the time period Natural gas unit price, Representative number is The nodes in the period The average cost correction item corresponding to the energy call frequency within .

[0035] As a further solution of the present invention, the specific steps of S5 are:

[0036] S501: Extract node records whose power path status has changed from the power supply level path table, locate the energy channel information under the corresponding node number, collect the activation and deactivation timestamp data of each channel, merge the time records by node number and sort them in chronological order to generate an energy channel activation and deactivation time series;

[0037] S502: Calling the energy channel on / off time sequence, calculating the adjacent on / off time values of the energy channel, obtaining the continuous activation time length of the channel, classifying and arranging the operation time of each channel by node number, and generating a channel continuous operation time result;

[0038] S503: According to the result of the continuous operation time of the channel, identify the node number whose continuous operation time is greater than the preset cycle lower limit, call the corresponding node number and operation time, channel type to merge and generate a path call information record table, and establish a virtual power plant adjustment plan.

[0039] As a further solution of the present invention, the calculation formula of the adjacent opening and closing time values of the energy channel is specifically:

[0040] ;

[0041] in, Represents the energy channel in The first closure and The adjacent opening and closing time value between openings, Representative channel The time of activation, Representative channel The time of closing, Representative channel The unit running weight of the segment, Representative The unit running weight of the segment, Representative The duration of the segment channel operation, Representative The duration of the segment channel operation, Representative Paragraph and Section The absolute value of the difference in running time between segments, Represents the weighted smoothing factor for the running difference correction term.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are:

[0043] In the present invention, by combining the direction of load and price changes, screening response deviation nodes, enhancing the strength of anomaly identification, utilizing the cross-judgment of temperature curve and energy cost difference, improving the rationality of path switching, introducing collaborative analysis of cost difference and spatial attributes, refining path attribution adjustment, generating path call information based on continuous activation time, supplementing the periodic tracking capability, and overall enhancing the dynamic adaptability and control accuracy of the regulation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 It is a schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0047] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0048] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0049] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0050] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0051] See also Figure 1 , a virtual power plant regulation method applicable to a medical system, comprising the following steps:

[0052] S1: Obtain the start and stop times of energy-using areas in medical scenarios, classify and mark the start and stop times by hour based on the time series, extract the time points of electricity price change nodes, compare the state changes with the electricity price nodes in chronological order, calculate the point-to-point interval length in the two time series, mark the types and operating trends within the interval, and generate a corresponding relationship table between operation and electricity price nodes;

[0053] S2: Based on the correspondence table between operation and electricity price nodes, extract nodes marked as load increase and decrease, obtain actual load data for the corresponding time period, compare the load change direction with the electricity price change trend, identify records with consistent and inconsistent directions through classification and comparison, separately file inconsistent records, and generate a response delay list;

[0054] S3: Extract the hot water heating node ID from the response delay list, obtain the time-divided temperature sequence of the hot water return end, obtain the cost data of electricity and natural gas prices in the same time period, calculate the cost difference between the two energy sources and perform a cross-judgment with the hot water temperature sequence. When both conditions meet the defined boundary interval, set the node heating path to the electric channel and generate an energy source conversion execution flag.

[0055] S4: Based on the node number in the energy source conversion execution tag, locate the lighting, electric cooling, and temperature control data in the space area, obtain the difference between the original energy call record and the energy cost in the current period, and determine whether the ownership change threshold is triggered. If triggered, classify the node power supply as a power path and update the level ownership status to generate a power supply level path table;

[0056] S5: Extract the node records after the change in the power supply level path table, obtain the activation and shutdown timestamps of the energy channel, calculate the running time of the channel continuously enabled, identify the path continuously calling the information file with the cycle time and the corresponding node number, and form a virtual power plant adjustment plan.

[0057] The correspondence table between operation and electricity price nodes includes start and stop time classification labels, electricity price node time series, comparison time labels of electricity price and operation status, operation interval type labels, and trend change types. The response delay list includes load change category, electricity price change direction, change direction consistency mark, direction inconsistent node number, and inconsistent record archiving label. The energy source conversion execution mark includes hot water heating node number, electricity and natural gas cost difference range, hot water temperature fluctuation range, and electricity channel switching mark. The power supply level path table includes node energy path classification, area identification code, lighting and temperature control category data, electricity cost difference range, and electricity path level mark. The virtual power plant adjustment plan includes node number index, energy channel operation start and end time, operation cycle duration, and path continuous call label.

[0058] The specific steps of S1 are:

[0059] S101: Obtain the start and stop times of energy usage areas in the medical scenario, divide the start and stop signals by hour, determine the relationship between the start and stop points and the corresponding hourly intervals, mark the operating status of each hour, and generate an hourly start and stop status type classification result;

[0060] To obtain the start and stop times of energy-using areas in medical scenarios, the monitoring data of each major energy-consuming equipment system in the hospital is extracted item by item. The process first retrieves the operation log data collected by the controllers of key energy-consuming areas such as air conditioning, fresh air system, lighting system, elevators and operating rooms. The data usually records the operating status of the equipment at the minute level. For example, "1" indicates running and "0" indicates stopping. The full-day data is then divided into 24-hour segments. Within each hour, the start and end marks of the start and stop status of the equipment within the hour are calculated. If the status of a device changes from 0 to 1 at 8:00, 8 o'clock is marked as the starting point. If the status changes from 1 to 0 at 17:00, 17 o'clock is marked as the stopping point. During the execution process, it is necessary to determine whether the start and stop points fall within the current hour range. For example, if the start point occurs at 8:12, it belongs to the 9th hour, that is, the 8:00-9:00 interval. This judgment logic is used to perform hour interval attribution matching for all start and stop time points. Based on this, an hourly operating status classification is constructed, and the following judgment is performed for each hour: if the hour contains a start point but no stop point, it is marked as a "start hour"; if it contains a stop point but no start point, it is marked as a "stop hour"; if there are no start or stop points but the equipment status is running, it is marked as a "continuous running hour"; if the status is stopped, it is marked as a "continuous stop hour". For example, if the lighting system is not running from 0:00 to 5:00, it is enabled at 5:00, and 5:00-6:00 is marked as a "start hour". It runs continuously from 6:00 to 22:00, marked as a "continuous running hour". If it stops at 22:00, 22:00-23:00 is marked as a "stop hour". In this way, a complete hourly start and stop status type classification result is constructed. The judgment logic for whether the start and stop points belong to a certain hour is realized by comparing whether the minutes of the start and stop time points are between the start and end times of the hour. For example, a start point at 7:45 is classified into the 7:00-8:00 hour period.

[0061] S102: Based on the hourly start / stop status classification results, detect the difference in electricity value at adjacent time points in the electricity price sequence, extract the change node, pair it with the hourly start / stop status, calculate the electricity price change correction value, and obtain the interval between the operating status and the electricity price node;

[0062] The calculation formula for the electricity price change correction value is as follows:

[0063] ;

[0064] in, Represents the electricity price change correction value of the current hour i, represents the electricity price value in the i-th hour, Indicates the electricity price value of the previous hour. represents the operating load value at hour i, represents the weight factor corresponding to the start-stop state in the i-th hour, It represents the average value of the operating load value in the continuous period related to the node. It represents the average value of the electricity price in the consecutive time periods related to the node;

[0065] parameter The electricity price in the ith hour is collected in real time by the electricity price interface of the energy consumption management system. The system is connected to the regional power grid dispatching platform and the collection frequency is once an hour. After monitoring and recording, the electricity price in the 9th hour is Yuan / kWh, electricity price for the previous hour Yuan / kWh.

[0066] parameter It represents the operating load value of the ith hour, which is obtained by the equipment monitoring system integrating the power data of the main electrical equipment in each area. The unit of this value is kW. According to actual statistics, the total load in the 9th hour is kW.

[0067] parameter The weight factor corresponding to the start and stop state in the i-th hour is set by the start and stop state sequence through the state mapping rule. The start state is assigned a value of 1.2, the stop state is assigned a value of 0.8, and the continuous operation state is assigned a value of 1.0. This mapping is set according to the degree of influence of the start and stop state on the load response sensitivity. When the state is "continuous operation", the load is stable and the sensitivity to changes is neutral, so it is set to 1.0. If the 9th hour is in the "start" state, then .

[0068] parameter is the average operating load of the current node for three consecutive hours. The load values in the 7th, 8th, and 9th hours are 138kW, 142kW, and 145kW respectively. The calculation results are:

[0069] ;

[0070] parameter The average electricity price for the corresponding three hours is 0.74, 0.76, and 0.88 yuan / kWh for the 7th, 8th, and 9th hours respectively. The calculation results are:

[0071] ;

[0072] Substituting the above values into the formula, the calculation process is as follows:

[0073] Calculate the absolute value of the electricity price difference:

[0074] ;

[0075] Calculate the load multiplied by the weight:

[0076] ;

[0077] Calculate the square root of the denominator:

[0078] ;

[0079] ;

[0080] ;

[0081] Finally, we substitute the formula to calculate:

[0082] ;

[0083] The result shows that the current electricity price change correction value for the 9th hour is 9.933. This value is obtained after comprehensive correction based on the start-stop behavior of each node and the load response weight. The larger the value, the greater the change amplitude under the current operating conditions. The value can be used to compare with the set threshold to determine whether the 9th hour is identified as an electricity price change node and participate in subsequent pairing and interval operations.

[0084] S103: Call the interval between the operating status and the electricity price node, compare the start and stop trend direction within the interval with the electricity price change direction, mark the time period status according to the trend consistency classification, and obtain the operating trend and electricity price node matching type table;

[0085] On the basis of matching the interval time data of the operating status and the electricity price node, the direction consistency of the operating trend and the electricity price change trend in each interval is compared and analyzed. When executing, the trend type of each start-stop change interval is first identified. Specifically, the trend from the previous starting point to the next stopping point is "start to stop", and vice versa is "stop to start". The trend direction is marked with the direction of change of the operating status value. For example, from state 0 to 1 is recorded as +1, and from 1 to 0 is recorded as -1. Similarly, the electricity price trend direction is extracted, such as from 0.75 yuan to 0.85 yuan is recorded as +1. If the operating trend direction is the same as the electricity price trend direction, it is considered to be consistent, otherwise it is considered inconsistent. When executing the judgment, the direction code of each start-stop trend interval is compared with the corresponding electricity price change interval. For example, in the air conditioner operating state from 8:00 to 12:00, the direction code of the start-stop trend interval is compared with the corresponding electricity price change interval. If the state changes from 0 to 1, recorded as +1, and the electricity price increases from 0.75 yuan to 0.85 yuan, recorded as +1, the trends are consistent. If the changes are in the opposite direction, the mark is inconsistent. The trend direction judgment does not introduce algorithm reasoning, and only uses the sign of the state value change to represent the trend direction. This method is applicable to all medical area equipment with clear operating status and electricity price time series data. For example, for the operating room lighting system that runs continuously from 12:00 to 18:00, if the electricity price shows a downward trend during this period, the operating trend is "continuous operation" and the electricity price trend is "downward". The two are inconsistent, so they are marked as inconsistent. Finally, by summarizing all trend matching results, a matching type table is formed, recording the start and end time of each interval, operating trend, corresponding electricity price change trend and consistency result for subsequent scheduling analysis.

[0086] The specific steps of S2 are:

[0087] S201: Based on the operation trend and electricity price node matching type table, filter the nodes marked as load increase and decrease, extract the time value of the corresponding time period, obtain the load collection sequence data within the time period, call the node time range and load collection sequence for interval matching, extract the load change of the node corresponding to the time period, and obtain a list of load change values;

[0088] First, a classification and filtering operation is performed on each node in the table. The filtering condition is the entry marked as "load increase" or "load reduction" in the node trend type field. When performing the filtering, the node trend field data is called, and the filtering value is set to the two text matching items of "load increase" and "load reduction". Only the nodes that meet the conditions are retained. Then the start and end time values of these nodes are extracted to form a time period set. For example, the start time of node N3 is 10:00 and the end time is 11:00, then the period is recorded as a 1-hour interval. Then, the load collection sequence data is called in the hospital energy consumption monitoring system. The sequence data is collected at a 5-minute granularity. For example, the load data from 10:00 to 11:00 is: 2.3kW, 2.5kW, 2.6kW, 2.8kW, 3.0kW, 3.2kW, 3.3kW, 3.4kW, 3.5kW, 3.6 kW, 3.7kW, 3.6kW, a total of 12 data points, each node time period is matched according to its corresponding start and end time, for example, the node N3 time period is matched to the load data of this section, and its load value at the start time is extracted as 2.3kW, and the load value at the end time is 3.6kW. The two are subtracted to obtain a load change value of +1.3kW. During the execution process, if there is an interruption or abnormal sampling point in the load sequence, such as the sampling missing time exceeds 10 minutes, the node will be invalidated and removed from participating in subsequent calculations. The judgment standard is set according to the maximum allowable missing rate of the hospital SCADA system not exceeding 15%. After executing the above process, the load change values corresponding to each node are summarized and organized uniformly to generate a list corresponding to the node number and the load change value, such as N1: +0.8kW, N2: -1.1kW, N3: +1.3kW, etc., which is used as the basis for subsequent judgment.

[0089] S202: Call the load change value list, classify and compare the load change direction and operation trend of each node with the corresponding electricity price change direction in the electricity price node matching type table, determine whether the directions are consistent, mark the consistent and inconsistent states according to the node number, and generate a load and electricity price trend consistency classification result;

[0090] Call the above-built load change value list, compare the load change direction corresponding to each node number with the operation trend and the electricity price change direction field corresponding to the node in the electricity price node matching type table, first classify and judge the load change direction. If the load change value is greater than 0, it is recorded as "increase", less than 0 is recorded as "decrease", and if it is equal to 0, it is recorded as "no change", such as N1: +0.8kW is "increase", N2: -1.1kW is "decrease", N4: 0 is "no change", and perform the same classification processing on the electricity price change direction field. If the electricity price rises from 0.75 yuan / kWh to 0.85 yuan / kWh, the change direction is "increase", if it drops from 0.85 yuan / kWh to 0.65 yuan / kWh, the change direction is "decrease", if not If there is a change, it is marked as "flat". Then the load direction and the electricity price direction are compared one by one. The comparison standard is: if the load is "increasing" and the electricity price is "rising" or the load is "decreasing" and the electricity price is "falling", the two directions are consistent, otherwise they are marked as inconsistent. For example, the load of node N1 increases and the electricity price rises, and the direction is consistent; the load of node N2 decreases and the electricity price rises, and the direction is inconsistent; the load of node N3 increases and the electricity price falls, and the direction is inconsistent. All nodes are judged and recorded according to this rule. If a node has "no change" in load or "flat" in electricity price, it is marked as "inconsistent" by default. This setting is to prevent misjudgment of consistency when the trend is unclear. Finally, the consistency classification result is formed according to the node number, such as N1: consistent, N2: inconsistent, N3: inconsistent, N4: inconsistent.

[0091] S203: Based on the load and electricity price trend consistency classification results, all nodes marked with inconsistent directions are extracted, and a corresponding list structure is established by calling the node number, load change value, and electricity price change direction. The nodes are sorted and archived by time to generate a response delay list;

[0092] According to the load and electricity price trend consistency classification results generated in the previous paragraph, extract the node number list of all direction inconsistent marks, such as N2, N3, and N4. In this number list, call the corresponding load value in the load change value list and the corresponding electricity price change direction field content in the operation trend and electricity price node matching type table for each node respectively, and build a combined structure of the three contents to form a list entry of node number, load change value, and electricity price direction, such as N2: -1.1kW, rising, N3: +1.3kW, falling, N4: 0, flat. The structure list is in a standard three-column format, followed by a node. The point time is the primary key field. The time information of each inconsistent node is retrieved from the operation trend and electricity price node matching type table. All entries are sorted in ascending order using the timestamp sorting method. For example, if the time of N3 is 9:00, N2 is 10:00, and N4 is 11:00, the sorted order is N3→N2→N4. After the sorting is completed, they are archived in order and a structured response delay list is output. All fields are classified into five columns according to node number, time, load change value, electricity price change direction, and inconsistency mark. For example, N3, 9:00, +1.3kW, decrease, inconsistent. This list is used for subsequent diagnosis of response time lag.

[0093] The specific steps of S3 are:

[0094] S301: Extract the hot water heating node numbers from the response delay list, obtain the corresponding time period for each number at the return end, collect the temperature series data of the hot water return end within the time period to which the node belongs, organize the correspondence between time and temperature by hour, establish the association structure between the hot water node and the time period temperature, and generate the hot water temperature series distribution result;

[0095] First, search the node records in the list one by one, and filter out the record items with the type mark of "hot water heating" according to the node classification field. This field is usually set to an enumeration value, and the matching items are "hot water heating", "air conditioning", "lighting", etc. The screening logic is to determine whether the type field in each record is equal to "hot water heating". Only the node number list that meets the conditions is retained, such as nodes H1, H3, and H7. Then, for each number, the return end time period recorded in the system is called. This time period is the hot water return monitoring time range of the heating system to which the node belongs. For example, the operating time period of node H1 is 8:00 to 10:00, then the corresponding return end temperature data needs to be extracted within this time period, and the return end temperature sequence recorded by the temperature acquisition system is called. This sequence generally records data with a granularity of 1 minute or higher. For example, the temperature sequence within 8:00-10:00 is 45 .2°C, 45.5°C, 46.0°C... etc. After obtaining all sampling points, the sequence is segmented by hour, that is, 8:00 to 9:00 is the first hour segment, and 9:00 to 10:00 is the second hour segment. The average temperature value of the temperature sampling values in the two hour segments is calculated as the representative temperature of the hour. If there are 60 temperature sample points in the first hour segment, the corresponding values are 45.0°C to 46.2°C, and the average is 45.6°C. The second hour segment is 46.3°C. The node and time period temperature association structure is constructed, that is, node H1: 8:00-9:00 is 45.6°C, 9:00-10:00 is 46.3°C. This is analogous to processing all hot water heating nodes. After completion, it is organized into a structure list consisting of three columns: hot water node number, time period, and average temperature, which is the hot water temperature sequence distribution result.

[0096] S302: Retrieving the hot water temperature sequence distribution results, collecting the unit price sequence data of electricity and natural gas costs within the same time period, calculating the cost difference by time, and combining the calculated result with the temperature value at the time point to determine whether it meets the preset cost difference threshold and temperature range limit, and generating the time node number that meets the energy switching judgment condition;

[0097] Call the above hot water temperature series distribution results, read the average temperature value of each node in a certain time period one by one, and call the electricity price series and natural gas price series data corresponding to the time period. For example, in the time period of 9:00-10:00, the electricity price is 0.85 yuan / kWh, the natural gas price is 2.4 yuan / Nm³, and the corresponding hot water temperature is 46.3°C. The cost difference between the two is calculated as the natural gas cost minus the electricity cost, and the difference of 1.55 yuan is obtained. This cost difference participates in the next judgment operation. The judgment step is to compare the cost difference and the temperature value to see whether they fall within the preset judgment threshold and temperature range. The cost difference threshold is set to 1.5 yuan, which is set based on the historical average of energy procurement. For example, the average monthly price difference between natural gas and electricity is 1.4 yuan. The upper limit plus the floating setting is 1.5 yuan. The temperature range limit is set to 45°C to 50°C. This range is the standard operating range between the lower limit of stable operation and the safety upper limit of the hospital hot water system. When judging, compare whether 1.55 yuan is greater than or equal to 1.5 yuan, and whether 46.3°C is between [45, 50]. If both are met, the node time period is marked as a time node that meets the switching conditions. If the cost difference is 1.2 yuan or the temperature is 44.8°C, the conditions are not met. The node time period numbers of all nodes that meet the above dual judgment criteria are uniformly recorded to form a time node number set as the basis for energy switching.

[0098] S303: Based on the time node number that meets the energy switching determination condition, set the heating path identification field according to the node number, set the heating path state to the electric channel state, and generate an energy source conversion execution flag;

[0099] Based on the list of time node numbers that meet the energy switching criteria obtained in the previous section, each number is read one by one and the corresponding hot water heating path status field is updated. Specifically, the structural configuration parameter table of the hot water heating node is called to find the corresponding heating path identification field. The original value may be "Gas Channel Status" or "Backup Channel Status". It is updated to "Electric Channel Status", indicating that the electric heating path should be used during this time period. The field update operation directly overwrites the original field value. The field content is an enumeration item-limited value. After the update is executed, a new column is added to the status flag field as an execution flag record. The value is "Conversion Executed" or "Not Executed" to distinguish whether the energy path conversion operation has been completed. For example, if node H1 meets the switching criteria during the time period of 9:00-10:00, its heating path field value is updated from "Gas Channel Status" to "Electric Channel Status", and the execution flag is set to "Conversion Executed". After completion, all node information is re-aggregated into a four-column structure consisting of node number, time period, new path status, and execution flag, forming the energy source conversion execution flag table.

[0100] The specific steps of S4 are:

[0101] S401: Based on the node number in the energy source conversion execution tag, locate the spatial area to which the corresponding node belongs, collect lighting equipment power data, cooling load records, and temperature control equipment switching time series in the area, merge the status records of all devices in the current period by node number, and generate a spatial area device status structure;

[0102] First, read the node list marked as "electrical channel status", for example, the node numbers are E01, E03, and E06, and perform spatial area positioning operations on these node numbers in turn, calling the floor, ward or functional unit number corresponding to the node in the building information model. For example, E01 corresponds to the second floor of the outpatient building, and E03 corresponds to the third floor of the inpatient building. Get the spatial area coordinates or logical partition identifier covered by each number, and then call the deployed lighting equipment monitoring system in the spatial area to extract all lamp power data. This data is a real-time power value recorded every 5 minutes. For example, the power of the lighting equipment on the second floor of the outpatient building is a sequence of 1.2kW, 1.3kW, 1.25kW, etc. At the same time, retrieve the cooling load data record of the electric cooling system in this area. For example, the cooling load between 9:00 and 10:00 is 35kW, 36.5kW, 37kW, etc. The on / off time series of the temperature control equipment is called, and the data records the start and stop timestamps of each temperature control equipment. For example, number T100 is turned on at 9:10 and turned off at 9:45, and number T102 is turned off at 9:00 and not enabled at 10:00. All equipment data is merged according to the node number, that is, within the time period of the node, the status information of all lighting, electric cooling, and temperature control equipment is integrated and reconstructed into a unified format according to their respective time axes. Finally, a multi-field combination structure containing node number, spatial area identifier, time period, lighting power sequence, electric cooling load sequence, and temperature control on / off time record is generated, that is, the spatial area device status structure.

[0103] S402: Call the spatial area device status structure, extract the energy call record corresponding to the node number, collect the unit price series of electricity and natural gas costs in the current period, calculate the difference between electricity and natural gas costs, and compare it with the ownership change cost threshold set for the node number to determine whether the change condition is triggered and generate the ownership change trigger status;

[0104] The formula for calculating the difference between electricity and natural gas costs is as follows:

[0105] ;

[0106] in, Representative number is The nodes in the period The difference between electricity and natural gas costs within Representative number is The nodes in the period The number of energy-related call devices, Representative number is The node of Devices in the time period The unit price of electricity within Representative number is The node of Devices in the time period Natural gas unit price, Representative number is The nodes in the period Average cost correction item corresponding to the energy call frequency within the period;

[0107] Parameter acquisition and value setting:

[0108] :Number is The nodes in the period The number of energy-related devices in the network. Statistical data is collected through the node device management system. .

[0109] :Number is The node of Devices in the time period The unit price of electricity within 2024. Based on the Brazilian commercial electricity price data in September 2024.

[0110] :Number is The node of Devices in the time period Natural gas unit price within 2024. Based on Brazilian commercial natural gas price data in September 2024.

[0111] :Number is The nodes in the period The average cost correction item corresponding to the energy call frequency within the period. Calculated based on the call frequency data recorded by the energy management system.

[0112] Formula calculation process:

[0113] Calculate the average unit price of electricity:

[0114] ;

[0115] Calculate the average unit price of natural gas:

[0116] ;

[0117] Calculate the cost difference:

[0118] ;

[0119] The results show that the number The nodes in the period The difference between electricity and natural gas costs is 0.072 reais.

[0120] S403: Based on the ownership change trigger status, filter the node IDs that have triggered the change condition, set the node power status to the power path, and update the ownership status of the level field to the power main path level, and establish a power supply level path table;

[0121] According to the trigger status of the ownership change, the node numbers with the trigger condition status of "yes" are filtered out, such as nodes E03, E05, and E08, and the corresponding energy configuration parameter records are read. The power status field is called according to the node number and set to "power path", indicating that the current energy ownership of the node uses the power path. Then the ownership status field content in the level field is updated. The original value of this field may be "backup path level" or "low priority channel", which is replaced with "power main path level". This assignment logic is to adjust the path that meets the energy switching economy judgment and the equipment operation is stable. After the field update is completed, the node number is used as the index to construct a structure combination containing the node number, power status field value, and level ownership field value. Finally, the power supply level path table is output by field sorting, forming a data field format of E03, power path, main path level and other data items for subsequent energy allocation.

[0122] The specific steps of S5 are:

[0123] S501: Extract node records whose power path status has changed in the power supply level path table, locate the energy channel information under the corresponding node number, collect the activation and deactivation timestamp data of each channel, merge the time records by node number and sort them in chronological order to generate the energy channel activation and deactivation time series;

[0124] First, read the path status field of all nodes in the table, compare the current field value with the previous archived field to see if it is consistent. If the field value is found to have changed from "gas path" or "backup path" to "power path", it is considered a node with a changed state. For example, node numbers D01, D04, and D09 meet the change condition. They are extracted and summarized to form a number set. Then, the energy channel configuration data corresponding to each node is read. The energy channel configuration data records the channel number and channel attributes attached to each node, mainly including fields such as power type, channel number, and channel control signal identification. For each channel, the start and close status identification sequence recorded by its control system is further extracted. The identification sequence is logical data, and the status is uploaded by the controller at regular intervals. For example, 0 means closed and 1 means open The reading order is reorganized according to the system number priority, and the state sequences of multiple channels under the same node are merged into a single data group. Then the state change points are logically sorted according to the channel number, that is, the turning position from "0" to "1" or "1" to "0" is found in each sequence, marked as the start and close event, and a change record of the logical sequence is formed. For example, the state sequence of node D01 channel T03 is [0, 0, 1, 1, 1, 0, 0, 1], then the start and close event positions correspond to enable → 3rd bit, close → 6th bit, and re-enable → 8th bit respectively. Finally, the state transition records of all channels under the node are merged and sorted by number to generate the energy channel start and close state sequence with the fields containing "node number, channel number, enable state, and enable sequence position".

[0125] S502: Call the energy channel on / off time sequence, calculate the adjacent on / off time values of the energy channel, obtain the continuous activation time length of the channel, classify and organize the operation time of each channel by node number, and generate the channel continuous operation time result;

[0126] The calculation formula for the adjacent opening and closing time values of the energy channel is as follows:

[0127] ;

[0128] in, Represents the energy channel in The first closure and The adjacent opening and closing time value between openings, Representative channel The time of activation, Representative channel The time of closing, Representative channel The unit running weight of the segment, Representative The unit running weight of the segment, Representative The duration of the segment channel operation, Representative The duration of the segment channel operation, Representative Paragraph and Section The absolute value of the difference in running time between segments, Represents the weighted smoothing factor for the running difference correction term;

[0129] The formula used to calculate the energy channel in The first closure and Enhanced time interval between openings .in,

[0130] :Channel No. The time of activation is obtained through the energy channel opening and closing time series data acquisition system, in hours.

[0131] :Channel No. The time of the first closing is obtained through the energy channel opening and closing time series data acquisition system, and the unit is hour.

[0132] 、 :Represents the channel Paragraph and Section The unit operation weight of the segment is calculated by the load monitoring system based on the load conditions of the channel during the operation of each segment, and the unit is dimensionless.

[0133] 、 :Represents the channel Paragraph and Section The actual running time of the segment is obtained from the running time monitoring system, in hours.

[0134] Set the specific values as follows:

[0135] Hour;

[0136] Hour;

[0137] , calculated by the load monitoring system, reflecting the channel in the first The load strength of the segment.

[0138] , calculated by the load monitoring system, reflecting the channel in the first The load strength of the segment.

[0139] hours, obtained through the runtime monitoring system, indicating that the channel is The running time of the segment.

[0140] hours, obtained through the runtime monitoring system, indicating that the channel is The running time of the segment.

[0141] Substituting the above values into the formula, the calculation process is as follows:

[0142] Step 1: Substitute the time difference and the weighted sum numerator and denominator

[0143] ;

[0144] Step 2: Calculate multiplication and denominator expansion

[0145] ;

[0146] Step 3: Sum and calculate division

[0147] ;

[0148] Step 4: Final calculation and absolute value

[0149] ;

[0150] Step 5: Get the final result value

[0151] ;

[0152] This result shows that the energy channel is The first closure and The enhanced interval between openings is 5.67973 hours. This value takes into account the difference in channel opening and closing times, operational load weight, and operating duration, reflecting the actual interval between adjacent channel openings and closings. This metric can be used to further analyze channel operational efficiency and optimize scheduling.

[0153] S503: Based on the channel continuous operation time result, identify the node number whose continuous operation time is greater than the preset cycle lower limit, call the corresponding node number, operation time, and channel type to generate a path call information record table, and establish a virtual power plant adjustment plan;

[0154] According to the channel continuous operation segment length record obtained in the previous paragraph, the "continuous segment length" field in each record is retrieved in turn, and compared with the set operation segment number threshold for judgment. The threshold is used to determine whether it has the adjustment capability. The threshold is set to 2 segments. This value is set according to the continuous response capability requirements of the typical electrical channel, that is, a channel with more than two consecutive enabled behaviors can be regarded as an adjustable channel. For each record, if the "continuous segment length" is ≥ 2 segments, the node number is recorded as the target node that meets the conditions. For example, the continuous segment of node D01 channel T03 is 3 segments, which meets the conditions. The node D05 channel T01 has only 1 segment, which does not meet the conditions. All node numbers that meet the conditions and their channel numbers and channel type fields (such as electrical channel, cold channel, etc.) are merged to generate record entries with the structure of "node number, channel number, channel type, continuous segment length", and uniformly output as a path call information record table. The table is sorted by node number and serves as the input data source for the subsequent virtual power plant adjustment logic.

[0155] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A virtual power plant adjustment method applicable to a medical system, characterized in that: The following steps are involved: S1: Obtain the start and stop times of the energy zones in the medical scenario, classify and mark them by hour, extract the electricity price change nodes, sequentially compare the operating status with the electricity price changes, calculate the point-to-point interval duration, and generate a corresponding relationship table between the operating and electricity price nodes; S2: Extracting load change nodes based on the table of correspondence between operation and electricity price nodes, obtaining actual load data for the corresponding period, comparing load and electricity price change directions, identifying consistent and inconsistent records, and generating a response delay list; S3: Extract the hot water node in the response delay list, obtain the return temperature sequence and the electricity and gas prices of the same period, determine whether the cost difference and temperature change meet the specified range, and if so, set it as an electricity channel and generate an energy source conversion execution flag; S4: Locate the node area based on the energy source conversion execution mark, extract lighting, electric cooling, temperature control data and original energy records, and determine whether to trigger a change based on the cost difference. If triggered, classify it as a power path and generate a power supply level path table; S5: Extract the change nodes in the power supply level path table, obtain the channel start and stop time, calculate the continuous operation time, identify the path call information in combination with the node number, and form a virtual power plant adjustment plan.

2. The virtual power plant adjustment method applicable to medical systems according to claim 1, characterized in that: The operation and electricity price node correspondence table includes start and stop time classification labels, electricity price node time series, comparison time length labels of electricity price and operation status, operation interval type labels, and trend change types. The response delay list includes load change categories, electricity price change directions, change direction consistency labels, direction inconsistent node numbers, and inconsistent record archiving labels. The energy source conversion execution mark includes hot water heating node numbers, electricity and natural gas cost difference ranges, hot water temperature fluctuation ranges, and electricity channel switching marks. The power supply level path table includes node energy path classifications, area identification codes, lighting and temperature control category data, electricity cost difference ranges, and electricity path level labels. The virtual power plant adjustment plan includes node number indexes, energy channel operation start and end times, operation cycle duration, and path continuous call labels.

3. The virtual power plant adjustment method applicable to medical systems according to claim 1, characterized in that: The specific steps of S1 are: S101: Obtain the start and stop times of energy usage areas in the medical scenario, divide the start and stop signals by hour, determine the relationship between the start and stop points and the corresponding hourly intervals, mark the operating status of each hour, and generate an hourly start and stop status type classification result; S102: Based on the hourly start / stop status type classification result, detect the difference in electricity value at adjacent time points in the electricity price sequence, extract the change node, pair it with the hourly start / stop status, calculate the electricity price change correction value, and obtain the interval between the operating state and the electricity price node; S103: Call the interval between the operating status and the electricity price node, compare the start-stop trend direction and the electricity price change direction within the interval, mark the time period status according to the trend consistency classification, and obtain the operating trend and electricity price node matching type table.

4. The virtual power plant adjustment method applicable to medical systems according to claim 3, characterized in that: The calculation formula for the electricity price change correction value is specifically as follows: ； in, Represents the electricity price change correction value of the current hour i, represents the electricity price value in the i-th hour, Indicates the electricity price value of the previous hour. represents the operating load value at hour i, represents the weight factor corresponding to the start-stop state in the i-th hour, It represents the average value of the operating load value in the continuous period related to the node. It represents the average value of the electricity price in consecutive time periods related to the node.

5. The virtual power plant adjustment method applicable to medical systems according to claim 3, characterized in that: The specific steps of S2 are: S201: Based on the operation trend and electricity price node matching type table, filter the nodes marked as load increase and decrease, extract the time value of the corresponding time period, obtain the load collection sequence data within the time period, call the node time range and the load collection sequence for interval matching, extract the load change of the node corresponding to the time period, and obtain a load change value list; S202: Calling the load change value list, classifying and comparing the load change direction and operation trend of each node with the corresponding electricity price change direction in the electricity price node matching type table, determining whether the directions are consistent, marking the consistent and inconsistent states according to the node number, and generating a load and electricity price trend consistency classification result; S203: According to the load and electricity price trend consistency classification results, extract all nodes marked with inconsistent directions, call the node number and load change value, and electricity price change direction to establish a corresponding list structure, sort and archive them by node time, and generate a response delay list.

6. The virtual power plant adjustment method applicable to medical systems according to claim 1, characterized in that: The specific steps of S3 are: S301: Extracting the hot water heating node numbers from the response delay list, obtaining the time period corresponding to each number at the return end, collecting temperature series data at the hot water return end within the time period to which the node belongs, arranging the correspondence between time and temperature by hour, establishing an association structure between the hot water node and the time period temperature, and generating a hot water temperature series distribution result; S302: The hot water temperature sequence distribution result is called, and the unit price sequence data of the electricity cost and the natural gas cost in the same time period are collected. The cost difference is calculated according to the time. The calculation result is combined with the temperature value at the time point to determine whether it meets the preset cost difference threshold and the temperature range limit, and the time node number that meets the energy switching determination condition is generated; S303: Based on the time node number that meets the energy switching determination condition, set the heating path identification field according to the node number, set the heating path state to the electric channel state, and generate an energy source conversion execution mark.

7. The virtual power plant adjustment method applicable to medical systems according to claim 1, characterized in that: The specific steps of S4 are: S401: Based on the node number in the energy source conversion execution tag, locate the spatial area to which the corresponding node belongs, collect lighting equipment power data, electric cooling load records, and switching time series of temperature control equipment in the area, merge the status records of all devices in the current time period according to the node number, and generate a spatial area device status structure; S402: Calling the spatial area device status structure, extracting the energy call record corresponding to the node number, collecting the unit price sequence of electricity and natural gas costs in the current period, calculating the difference between the electricity and natural gas costs, and comparing it with the ownership change cost threshold set for the node number to determine whether the change condition is triggered and generate the ownership change trigger status; S403: According to the ownership change trigger status, filter the node numbers that have triggered the change conditions, set the node power status to the power path, and update the ownership status of the level field to the power main path level, and establish a power supply level path table.

8. The virtual power plant adjustment method applicable to medical systems according to claim 7, characterized in that: The formula for calculating the difference between electricity and natural gas costs is as follows: ； in, Representative number is The nodes in the period The difference between electricity and natural gas costs within Representative number is The nodes in the period The number of energy-related call devices, Representative number is The node of Devices in the time period The unit price of electricity within Representative number is The node of Devices in the time period Natural gas unit price, Representative number is The nodes in the period The average cost correction item corresponding to the energy call frequency within .

9. The virtual power plant adjustment method applicable to medical systems according to claim 1, characterized in that: The specific steps of S5 are: S501: Extract node records whose power path status has changed from the power supply level path table, locate the energy channel information under the corresponding node number, collect the activation and deactivation timestamp data of each channel, merge the time records by node number and sort them in chronological order to generate an energy channel activation and deactivation time series; S502: Calling the energy channel on / off time sequence, calculating the adjacent on / off time values of the energy channel, obtaining the continuous activation time length of the channel, classifying and arranging the operation time of each channel by node number, and generating a channel continuous operation time result; S503: According to the result of the continuous operation time of the channel, identify the node number whose continuous operation time is greater than the preset cycle lower limit, call the corresponding node number and operation time, channel type to merge and generate a path call information record table, and establish a virtual power plant adjustment plan.

10. The virtual power plant adjustment method applicable to medical systems according to claim 9, characterized in that: The calculation formula for the adjacent opening and closing time values of the energy channel is specifically as follows: ； in, Represents the energy channel in The first closure and The adjacent opening and closing time value between openings, Representative channel The time of activation, Representative channel The time of closing, Representative channel The unit running weight of the segment, Representative The unit running weight of the segment, Representative The duration of the segment channel operation, Representative The duration of the segment channel operation, Representative Paragraph and Section The absolute value of the difference in running time between segments, Represents the weighted smoothing factor for the running difference correction term.

Images