Method for separating legal metering software of single-chip gas meter

Abstract

The invention relates to the technical field of gas data processing, in particular to a method for separating legal metering software of a single-chip gas meter, which comprises the following steps of: 1, processing historical metering data in a legal metering area, and determining characteristic data; 2, calculating a feature smoothing function based on the feature data; step 3, determining a curve change function according to the time period response time and the time period data, setting a change threshold value and taking the change threshold value as an input value, and obtaining a data node value; 4, determining an estimated value of the current unit time, comparing the estimated value with a data node value, and determining a main energy consumption function of the legal metering area; by setting the legal metering area and the illegal metering area, legal metering data are analyzed in the legal metering area, supervision of the legal metering data is facilitated, meanwhile, historical metering data in the legal metering area are analyzed, and then main energy consumption functions in the single-chip gas meter are reasonably distributed. And the response function of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas data processing, and particularly to a method for separating the legal metering software of a single-chip gas meter. Background Art

[0002] With the development of integrated circuit technology and information technology, gas meters are gradually developing towards the intelligent direction. The rise of the Internet of Things technology has promoted the further upgrading of gas meter technology. The Internet of Things intelligent gas meter has become one of the mainstream products in the market. It can not only realize functions such as remote meter reading, remote control, and online billing, but also perform data interaction with the management system of the gas company to achieve intelligent gas management.

[0003] The prior art CN115904410A discloses a method for separating the legal area and the illegal area of an intelligent electric meter, including: Step S1: Divide the overall software architecture of the intelligent electric meter into three parts: BootLoader, legal area, and illegal area; Step S2: Perform multi-project link configuration on BootLoader, the legal area, and the illegal area; Step S3: Partition the Flash area and the RAM area of each project respectively.

[0004] However, when separating the legal metering software of a gas meter, since the flow data generated by the gas is different at different time periods, and the amount of data to be processed is also different during legal metering. When the gas meter is in the peak period and the low peak period of the flow data respectively, the data response time is different. At this time, when processing the legal metering data in a single chip, if multiple functions in the legal metering area are processed in parallel, the various energy consumptions in the single chip occupy a large amount of resources. Especially in the peak period of the flow data, it will cause the data quality in the legal metering area of the gas meter to decline, thereby reducing the metering accuracy of the gas meter. Summary of the Invention

[0005] The purpose of the present invention is to solve the problems in the background art, and to propose a method for separating the legal metering software of a single-chip gas meter.

[0006] To achieve the above purpose, the present invention adopts the following technical solution:

[0007] A method for separating the legal metering software of a single-chip gas meter, the method specifically includes the following steps:

[0008] Step 1: Set the legal metrology area and the non-legal metrology area in the software area of the gas meter, obtain the historical metrology data of the legal metrology area of the gas meter, divide the historical metrology data according to the unit time to obtain the period data of the unit time. At the same time, set the cycle period for the unit time, set the position number for the unit time in the cycle period, extract the period data with the same position number, and calculate the mean value to obtain the characteristic data corresponding to the position number;

[0009] Step 2: Arrange the characteristic data in the order of the position numbers in the cycle period. At the same time, use the characteristic data as input data, use the neural network algorithm to determine the characteristic smoothing coefficient α, and determine the characteristic smoothing function based on the characteristic smoothing coefficient;

[0010] Step 3: Divide the data response time according to the unit time to obtain the period response time. Mark the period response time and the period data of each unit time in the plane coordinate system, and perform curve fitting to generate a curve function. Calculate the first derivative of the curve function to obtain the curve change function. Then set the change threshold and input it into the curve change function to obtain the data node value;

[0011] Step 4: Obtain the estimated value at the previous position number and the actual generated period data as input data, and input them into the value characteristic smoothing function to obtain the estimated value corresponding to the current unit time. Compare the estimated value corresponding to the current unit time with the data node value to determine the main energy consumption function of the legal metrology area.

[0012] As a further solution of the present invention, the method for determining the characteristic data of the position number includes:

[0013] S1: Take the current time as the time node, obtain the historical metrology data of the legal metrology area, set the unit time, divide the historical metrology data according to the unit time, and mark the data volume in each unit time as the period data;

[0014] S2: Select consecutive n unit times as the cycle period. At this time, divide the historical time corresponding to the historical metrology data into multiple cycle periods, where the value of n is set to 12, and one cycle period is one natural day;

[0015] S3: Arbitrarily select a position number and use it as the target number. In all cycle periods, extract the period data corresponding to the target number position and mark it as DTj, where j represents different cycle periods;

[0016] After that, use the normal distribution algorithm to identify the abnormal data and normal data in the period data DTj;

[0017] S4: Extract the normal data from the time period data DTj, calculate the mean of the normal data in the time period data DTj, and mark the result of the mean calculation as the characteristic data at the target number position.

[0018] As a further solution of the present invention, the method for identifying abnormal data and normal data includes:

[0019] Extract the mean from the time period data DTj and mark it as DTa, and then use the formula to obtain the standard deviation μ of the time period data, where J represents the total number of cycle periods;

[0020] Based on the standard deviation μ and the mean DTa of the time period data, take [DTa - kμ, DTa + kμ] as the normal distribution interval, and the value of k is set to 2;

[0021] Compare the time period data DTj at the target number position with the normal distribution interval [DTa - kμ, DTa + kμ]. If DTj ∈ [DTa - kμ, DTa + kμ], mark the corresponding time period data as normal data. Otherwise, if then mark the corresponding time period data as abnormal data.

[0022] As a further solution of the present invention, the method for determining the characteristic smoothing function includes:

[0023] Arrange the characteristic data in the order of the position numbers in the cycle period. At the same time, use the characteristic data as the input data, determine the characteristic smoothing coefficient α using an intelligent algorithm, and determine the characteristic smoothing function based on the characteristic smoothing coefficient: Si = αDT i-1 +(1 - α)S i-1 ;

[0024] where Si represents the estimated value at the position number i, the smoothing index α ∈ (0, 1), S i-1 is the estimated value at the position number (i - 1), and DT i-1 is the actually generated time period data at the position number (i - 1).

[0025] As a further solution of the present invention, when the value of i is 1, obtain the unit time corresponding to the last position number in the previous cycle period at this time, and use the data corresponding to the last position number in the previous cycle period to replace the estimated value S i-1 and the actually generated time period data DT i-1 at the position number i - 1.

[0026] As a further solution of the present invention, the method for determining the data node value includes:

[0027] Divide the data response time by unit time and label the division result as the period response time. At this time, each period of data corresponds to a period response time.

[0028] Set the period data as the abscissa and the period response time as the ordinate, set up a plane coordinate system, and then mark the period data and period response time in each unit time in the plane coordinate system to obtain a scatter plot.

[0029] Use the Maltab tool to fit the scatter plot into a curve and generate a curve function Ft.

[0030] Perform a first derivative operation on the curve function Ft to obtain a curve change function L = Ft', where L represents the curve change rate.

[0031] Set a change threshold Ly, substitute the change threshold Ly for the curve change rate L into the curve change function for calculation, and output the corresponding period data. At this time, the output period data is the data node value.

[0032] As a further solution of the present invention, the method for determining the main energy consumption function in the legal metrology area includes:

[0033] Based on the feature smoothing function, obtain the estimated value and the actual generated period data at the previous position number as input data, and input them into the feature smoothing function to obtain the estimated value Si corresponding to the current unit time.

[0034] Compare the estimated value Si corresponding to the current unit time with the data node value. If the estimated value Si ≤ the data node value, set the metrological calibration function as the main energy consumption function within the current unit time. Otherwise, if the estimated value Si > the data node value, set the data processing function as the main energy consumption function within the current unit time.

[0035] Compared with the existing technology, the advantages of the present invention are:

[0036] By setting up a legal metrology area and an illegal metrology area, the present invention analyzes the legal metrology data in the legal metrology area, which is conducive to the supervision of legal metrology data, and then optimizes the software design, further improving the data processing efficiency.

[0037] The present invention analyzes the historical measurement data in the legal metrology area, estimates the period data within the current unit time, comprehensively analyzes the period data and the period response time in the unit time to determine the data node value, compares the data node value with the estimated value of the current unit time to determine the main energy consumption function of the legal metrology area, and then reasonably allocates the main energy consumption functions in the single-chip gas meter, improves the response function of the system, balances the functional energy consumption in the single-chip gas meter, reduces the additional loss, and thus prolongs the overall service life of the single-chip gas meter. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic structural diagram of the method flow of the present invention.

[0039] Figure 2 It is a schematic diagram of the legal metrology software flow. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0041] Refer to Figure 1 , a method for separating the legal metrology software of a single-chip gas meter, which specifically includes the following steps:

[0042] Embodiment 1:

[0043] Step 1: Set the legal metrology area and the illegal metrology area in the software area of the gas meter;

[0044] Identify the legal metrology area of the gas meter, monitor the flow data of each period in the legal metrology area, and analyze the historical measurement data at the same time to determine the characteristic data of each unit time. The measurement area refers to the area where an enterprise centrally manages the gas meters. The specific method for determining the characteristic data includes:

[0045] S1: Take the current time as the time node, obtain the historical measurement data of the legal metrology area, set the unit time, divide the historical measurement data according to the unit time, and mark the data volume in each unit time as the period data;

[0046] It should be further noted that the period data here refers to the total data volume collected by the gas meter for the measurement data within the unit time. The specific value of the unit time is set according to the big data experience of those skilled in the art. Among them, the unit time in this embodiment is set to 2 hours;

[0047] S2: Select consecutive n unit times as the cycle period. At this time, divide the historical time corresponding to the historical measurement data into multiple cycle periods;

[0048] Specifically, the value of n is set by those skilled in the art according to big data experience. In this embodiment, the value of n is set to 12, that is, a cycle period is a whole day. Further, a cycle period is a natural day;

[0049] After that, arrange the time period data in the cycle period in chronological order, and set a position number i for the position of the arranged time period data, where i = 1, 2,..., n;

[0050] S3: Arbitrarily select a position number as the target number. In all cycle periods, extract the time period data corresponding to the target number position respectively, and mark it as DTj, where j represents different cycle periods;

[0051] After that, use the normal distribution algorithm to identify the abnormal data in the time period data DTj. The specific method for identifying abnormal data includes:

[0052] First, take the mean value in the time period data DTj and mark it as DTa, and then use the formula to obtain the standard deviation μ of the time period data, where J represents the total number of cycle periods;

[0053] It should be further noted that the principle and calculation process of the normal distribution algorithm are both prior arts, so they will not be elaborated here;

[0054] According to the standard deviation μ and the mean value DTa of the time period data, take [DTa - kμ, DTa + kμ] as the normal distribution interval, k ∈ (0, 3]. In this embodiment, the value of k is set to 2;

[0055] Compare the time period data DTj at the target number position with the normal distribution interval [DTa - kμ, DTa + kμ]. If DTj ∈ [DTa - kμ, DTa + kμ], then mark the corresponding time period data as normal data. On the contrary, if then mark the corresponding time period data as abnormal data;

[0056] S4: Take the normal data in the time period data DTj, and calculate the mean value of the normal data in the time period data DTj, and mark the mean calculation result as the characteristic data at the target number position;

[0057] After that, sequentially use the remaining position numbers in the cycle time as the target numbers, obtain the time period data corresponding to the target number positions in each cycle period, and then process them according to the method in step S3 above, so as to obtain the characteristic data of each position number;

[0058] Step 2: Arrange the feature data in the order of the position numbers in the cycle period. At the same time, take the feature data as input data, use an intelligent algorithm to determine the feature smoothing coefficient α, and determine the feature smoothing function based on the feature smoothing coefficient: Si = αDT i-1 +(1 - α)S i-1 ;

[0059] where Si represents the estimated value at position number i, the smoothing exponent α ∈ (0, 1), S i-1 is the estimated value at position number (i - 1), and DT i-1 is the actual generation period data at position number (i - 1);

[0060] It should be further noted that when i takes the value of 1, obtain the unit time corresponding to the last position number in the previous cycle period at this time, and replace the estimated value S i-1 and the actual generation period data DT i-1 at position number i - 1 with the data corresponding to the last position number in the previous cycle period;

[0061] wherein, the intelligent algorithm in this embodiment is selected as the neural network algorithm, and the specific processing process of the neural network algorithm is the prior art and will not be elaborated here;

[0062] Step 3: Obtain the historical measurement data in the legal metrology area of the legal metrology software of the single chip, and at the same time obtain the data response time corresponding to the historical measurement data. According to the data response time of the historical measurement data, determine the data node value of the data volume. The specific method for determining the data node value includes:

[0063] First, divide the data response time by unit time, and mark the division result as the period response time. At this time, each period data corresponds to a period response time;

[0064] It should be further noted that the period response time is the average value of the data response times corresponding to all measurement data in the corresponding unit time. For example, in the unit time, there are measurement data b1, b2, b3, and b4, where the response time of b1 is detected as t1, the response time of b2 is t2, the response time of b3 is t3, and the response time of b4 is t4. At this time, the period response time is (t1 + t2 + t3 + t4) / 4;

[0065] After that, set the period data as the abscissa, set the period response time as the ordinate, set up a plane coordinate system, and then mark the period data and the period response time in each unit time in the plane coordinate system to obtain a scatter plot;

[0066] Use the Maltab tool to fit the scatter plot into a curve and generate the curve function Ft;

[0067] Perform a first derivative operation on the curve function Ft to obtain the curve change function L = Ft', where L represents the curve change rate;

[0068] Set the change threshold Ly, substitute the change threshold Ly for the curve change rate L into the curve change function for calculation, and output the corresponding time period data. At this time, the output time period data is the data node value;

[0069] It should be further noted that the specific value of the change threshold Ly is obtained by those skilled in the art through big data operations, and the curve change rate represents the influence factor of the time period data on the time period response time. When the curve change rate is larger, it means that the time period data has a greater impact on the time period response time;

[0070] Step 4: Based on the feature smoothing function, obtain the estimated value and the actually generated time period data at the previous position number as input data, and input them into the feature smoothing function to obtain the estimated value Si corresponding to the current unit time;

[0071] After that, compare the estimated value Si corresponding to the current unit time with the data node value. If the estimated value Si ≤ the data node value, then set the metering calibration function as the main energy consumption function within the current unit time. Conversely, if the estimated value Si > the data node value, then set the data processing function as the main energy consumption function within the current unit time;

[0072] It should be further noted that the metering calibration function refers to the latest legal metrology standards and requirements, and calibrates the legal metrology data to ensure the compliance of metering. The data processing function refers to functions such as data collection and compensation processing of legal metrology data;

[0073] Estimating the data volume of the current unit time can balance the functional power consumption in a single chip, and at the same time optimize the system's response ability, thereby ensuring the accuracy and reliability of the legal metrology part of the data after the separation of the legal metrology software in the gas meter, and further guaranteeing the data quality of the legal metrology data of the gas meter.

[0074] Embodiment 2: The legal metrology software part (LMSM) is separated from the control function software part (CFSM). It is required that the single-chip microcomputer divides different program areas and data areas for the software program, and the two parts of the program need to interact through a specific software interface. The recommended overall structure of the software program, data, and interaction is as Figure 2 shown;

[0075] Overall, the working process of the entire device is controlled by the Legal Metrology Software Module (LMSM). The Control Function Software Module (CFSM) is only called by the LMSM for specific events. Among them, for some interrupt function processes, the LMSM calls the hook function of the CFSM when needed, and after the call, it returns to the LMSM for subsequent processing or interrupt return; in the process of handling certain events, the control function function interface of the CFSM can be called to complete the corresponding functions; in the functions of the CFSM, if necessary, the callable functions opened by the LMSM can be called to complete the corresponding processing process.

[0076] The software interaction between the LMSM and the CFSM is carried out through a shared data / variable buffer set at a specific address in the SRAM. When the call process involves metrological data or sensitive data, encryption or encrypted verification should be adopted to ensure the legality of the calling party. Generally, when the LMSM is powered on for the first time, it should verify the identity information and software version of the CFSM by calling the corresponding functions of the CFSM to avoid malicious software attacks or data compatibility problems.

[0077] Step 1: Power on the LMSM program

[0078] Initialize the memory and hardware, and start the watchdog;

[0079] Sample and wait for the user's battery voltage to stabilize;

[0080] Read and verify data from the FLASH;

[0081] Enable interrupts and enable wake-up sources;

[0082] Conduct a full LCD self-check and display the LMSM software logo;

[0083] Verify the identity and version information of the CFSM;

[0084] Enter the main program loop, mark various events, handle events, and switch between sleep and wake-up.

[0085] Step 2: Sampling event

[0086] Anti-interference processing for pulse meter sampling;

[0087] The pulse meter calculates the pulse interval (instantaneous flow rate);

[0088] The pulse meter performs pulse counting (volume accumulation) for metering;

[0089] The ultrasonic meter collects flow velocity or volume for metering;

[0090] Call the CFSM to execute the extended functions for processing metering results (deduction, flow control, etc.).

[0091] Step 3: Timer event

[0092] Calculate and update the extended timer event;

[0093] Regularly collect and monitor the battery voltage;

[0094] Regularly collect the data of the temperature and pressure sensor (if any);

[0095] Call the CFSM to execute the timing extension function.

[0096] Step 4: Low-power timer event (can wake up from sleep)

[0097] Timekeeping and clock processing;

[0098] Generate clock timing events such as hours / days, etc.

[0099] Call the CFSM to execute the clock timing extension function (such as communication / hour / day / month usage record)

[0100] Step 5: Button event (can wake up from sleep)

[0101] Button anti-shake processing;

[0102] Call the CFSM to execute the button function operation;

[0103] Step 6: Power-off event

[0104] Save the metering / core data, close the valve, and give an LCD prompt;

[0105] Call the CFSM to save the extended data;

[0106] Close other wake-up sources, only keep the button wake-up, and enter sleep.

[0107] Meanwhile:

[0108] The legal metrology software part and the legal metrology related parameters (including verification information) are not allowed to use off-chip memory. The on-chip memory should be set with hardware / software control to avoid unauthorized reading or tampering.

[0109] The legal metrology software part uses CRC16 verification. Its verification code and software version should be displayed on the LCD when powering on or when necessary. The CRC16 verification algorithm uses the initial value 0x0000; the polynomial coefficient 0x1021; and the data processing method without exclusive OR of the result.

[0110] The legal metrology related parameters (including verification information) and the legal metrology core operation data should be stored in an encrypted manner or use encrypted data verification to avoid unexpected use and modification.

[0111] Data encryption uses the standard AES algorithm with a 128-bit key.

[0112] Encryption verification needs to introduce methods such as random numbers or the unique ID of the chip to improve the anti-attack performance.

[0113] As described above, it is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered within the protection scope of the present invention.

Claims

1. A method for separating legal measurement software of a single-chip gas meter, characterized in that: The method specifically comprises the following steps: Step 1: Set the legal metering area and the illegal metering area for the software area of ​​the gas meter, obtain the historical metering data of the legal metering area of ​​the gas meter, divide the historical metering data according to the unit time, obtain the time period data of the unit time, set the cycle period for the unit time, set the position number for the unit time in the cycle period, extract the time period data of the same position number, calculate the mean, and obtain the characteristic data of the corresponding position number; Step 2: Arrange the feature data in the order of the position numbers in the cycle, use the feature data as input data, use the neural network algorithm to determine the feature smoothing coefficient α, and determine the feature smoothing function based on the feature smoothing coefficient; Step 3: Divide the data response time according to the unit time to obtain the period response time, mark the period response time and period data of each unit time in the plane coordinate system, and perform curve fitting to generate a curve function, perform a derivative calculation on the curve function to obtain the curve change function, then set the change threshold and input it into the curve change function to obtain the data node value; Step 4: Get the estimated value at the previous position number and the actual time period data as input data, and input them into the value characteristic smoothing function to obtain the estimated value corresponding to the current unit time, compare the estimated value corresponding to the current unit time with the data node value, and determine the main energy consumption function of the legal metering area.

2. The method for separating legal measurement software of a single-chip gas meter according to claim 1, characterized in that: Methods for determining characteristic data of position numbers include: S1: Take the current time as the time node, obtain the historical measurement data of the legal measurement area, set the unit time, divide the historical measurement data according to the unit time, and mark the data volume in each unit time as time period data; S2: Select n consecutive unit times as the cycle period. At this time, the historical time corresponding to the historical measurement data is divided into multiple cycle periods, where the value of n is set to 12, and one cycle period is one natural day; S3: randomly select a position number as the target number, extract the time period data corresponding to the position of the target number in all cycles, and mark them as DTj, where j represents different cycles; Then, the normal distribution algorithm is used to identify abnormal data and normal data in the time period data DTj; S4: Take the normal data in the time period data DTj, perform mean calculation on the normal data in the time period data DTj, and mark the mean calculation result as the feature data at the target number position.

3. The method for separating legal measurement software of a single-chip gas meter according to claim 2, characterized in that: Methods for identifying abnormal data and normal data include: Take the mean of the period data DTj and mark it as DTa, then use the formula Get the standard deviation μ of the period data, where J represents the total number of cycles; According to the standard deviation μ and the mean DTa of the period data, [DTa-kμ, DTa+kμ] is taken as the normal distribution interval, and the value of k is set to 2; Compare the time period data DTj at the target number position with the normal distribution interval [DTa-kμ, DTa+kμ]. If DTj∈[DTa-kμ, DTa+kμ], the corresponding time period data is marked as normal data. Otherwise, The corresponding time period data is marked as abnormal data.

4. The method for separating legal measurement software of a single-chip gas meter according to claim 1, characterized in that: The method for determining the characteristic smoothing function includes: The characteristic data are arranged in the order of the position number in the cycle, and the characteristic data are used as input data, and the characteristic smoothing coefficient α is determined by an intelligent algorithm, and the characteristic smoothing function is determined based on the characteristic smoothing coefficient: Si = αDT i-1 +(1-α)S i-1 ; Where Si represents the estimated value at position number i, the smoothing index α∈(0,1), S i-1 is the estimated value at position number (i-1), DT i-1 It is the actual generation period data at position number (i-1).

5. The method for separating legal measurement software of a single-chip gas meter according to claim 4, characterized in that: When i is 1, the unit time corresponding to the last position number in the previous cycle is obtained, and the estimated value S at the position number i-1 is i-1 and the actual generation period data DT i-1 Replace it with the data corresponding to the last position number in the previous cycle.

6. The method for separating legal measurement software of a single-chip gas meter according to claim 1, characterized in that: Methods for determining data node values ​​include: The data response time is divided into units of time, and the division result is marked as a time period response time. In this case, each time period data corresponds to a time period response time. The time period data is set as the horizontal axis, the time period response time is set as the vertical axis, a plane coordinate system is set, and then the time period data and the time period response time in each unit time are marked in the plane coordinate system to obtain a scatter plot; Use the Maltab tool to fit the scatter plot into a curve and generate the curve function Ft; Perform a derivative operation on the curve function Ft to obtain the curve change function L = Ft', where L represents the curve change rate; Set the change threshold Ly, and input the change threshold Ly instead of the curve change rate L into the curve change function for calculation, and output the corresponding time period data. At this time, the output time period data is the data node value.

7. The method for separating legal measurement software of a single-chip gas meter according to claim 1, characterized in that: Methods for determining the main energy consumption functions in the legal measurement area include: Based on the characteristic smoothing function, the estimated value at the previous position number and the actual generation period data are obtained as input data, and input into the characteristic smoothing function to obtain the estimated value Si corresponding to the current unit time; Compare the estimated value Si corresponding to the current unit time with the data node value. If the estimated value Si ≤ the data node value, the measurement calibration function is set to the main energy consumption function in the current unit time. Conversely, if the estimated value Si > the data node value, the data processing function is set to the main energy consumption function in the current unit time.

Images