RU2528137C2 - Method for automated warning of fire caused by arcing in power grids or electrical installations and apparatus therefor - Google Patents

Abstract

FIELD: chemistry.

SUBSTANCE: invention relates to fire safety and specifically to methods and apparatus for warning fires caused by faults in power grids or electrical installations in rooms, buildings, aircraft, ships, railway transport etc. Low-frequency components (<10³ kHz) are suppressed in measured resultant current flowing in a protected circuit. Without storage, the value of the obtained signal is compared with a given value and a pulse with a fixed duration which is slightly less than half the period of the mains supply is generated every time said value is exceeded. The pulse is transmitted to a storage unit with a charging time constant of 0.1…0.3 s and a discharge time constant of about 0.4…0.6 s. The signal is transmitted from the output of the storage unit to the input of an additional comparator unit where it is compared with a given second level value. If said second level value is exceeded, a voltage jump occurs at the output of the comparator unit, which triggers a command generating unit which generates a pulse which is transmitted to an actuating device which disconnects the protected circuit from the mains and (or) outputs a warning signal.

EFFECT: high fire safety.

2 cl, 1 dwg

Description

The invention relates to the field of fire safety and electric power, and in particular to methods and devices for preventing fires arising from malfunctions in electrical networks or electrical installations in rooms, structures, buildings, aircraft, ships, railway transport and other objects.

The operation of residential, domestic, industrial and other facilities is often accompanied by fires that occur due to malfunctions in the wiring or other elements of the electrical network and electrical installations. As a result, great material and moral damage is inflicted, often accompanied by the death and (or) injury of people.

A large proportion of them (up to 80%) is occupied by fires that arise due to the lack of methods for detecting malfunctions such as sparking caused by poor-quality installation of elements of electrical networks and electrical installations, violations of their operation requirements and other reasons.

The nature of sparking is explained by the formation of ionized transient resistance in places of insufficiently close contact between movable and fixed contacts of switching elements (relays, switches, disconnectors, etc.), as well as in places of poor-quality connection of wires, buses, feeders (junction boxes, plug and other connections) and connecting to the power grid various elements (electric lamps, fuses, etc.). The dynamics of the transition resistance determines the process of periodic manifestation of the cycle "occurrence-suppression" of a spark (electric arc), accompanied by the release of thermal energy.

A technical solution is known [1], according to which the total electric current is measured at the input of an electric network or electrical installation. From the measured total current by filtering the first harmonic or the entire low-frequency spectrum, a signal of the second and / or higher harmonics is determined, which is determined by the transition resistance R _junction . The generated signal is amplified, rectified and accumulated over a set time. During the accumulation process, the magnitude of the total signal is compared with a predetermined value or with predetermined values of the first or higher levels of comparison. Each comparison level is selected for the corresponding degree of fire hazard of the electrical network or electrical installation (EU). Depending on the magnitude of the accumulated signal corresponding to the achieved comparison level, a signal or warning signals are generated about the occurrence of a fire hazard and / or a command is issued to turn off the electrical network or electrical installation (EU).

A significant drawback of this technical solution is the validity of its operation algorithm only with the active nature of the EU impedance in the protected circuit (heating devices, incandescent lamps, linear secondary power sources, etc.). When the electrical network (electrical installation) is in good condition, a signal whose frequency is equal to the frequency of the supply network is almost completely filtered out and at the input of the amplification unit, regardless of the magnitude of the load current, in the absence of a fault, the signal value is always close to zero. Therefore, after rectification and accumulation units, a constant component will take place, the value of which is determined by the filter tuning error, nonlinear distortions of the measured load current and the admissible transition resistance values in each of the contacts of the switching equipment and wiring and is perceived by the comparison unit as a logical zero [1].

However, at present, in residential, administrative, and industrial facilities, the majority of EIs have pulsed power sources - IIPs (computers, televisions, office equipment, etc.), the spectrum of pulsed current consumption of which includes intense harmonic components in the range of at least tens of kilohertz. Therefore, firstly, the average rectified value of the high-frequency components of the current sensor signal used in the prototype as a criterion for the health of the circuit will be non-zero, and secondly, its constancy will be violated when each pulse load is connected or disconnected (turning on the TV, computer, their transition into sleep mode), perceived by the system as arcing. On the other hand, intense sparking in a circuit even of a heating (impedance-active) power plant with a sufficiently high power, for example 1 kW, can be quite a long time in the nature of an almost periodic process with a fundamental frequency of 100 or 50 Hz, maintaining a constant mean-straightened value of the high-frequency components of the signal at intense heating the contact and flying sparks not detected by the system. Perhaps that is why not a single device has yet been created that, instead of signaling at its output, would produce a real automatic shutdown of the protected circuit during sparking, without which a wide and universal application of such a device is impossible.

The objective of the invention is the prevention of fires that occur in rooms, structures, buildings. The technical result is to increase fire safety.

To achieve a technical result in the claimed invention, the signal of the sensor, which is a current transformer consumed by the protected circuit, passes the RF spectrum extraction unit, which suppresses spectral components with a frequency below several hundred kilohertz. This makes it possible to discriminate in amplitude the signals of the currents of pulsed power supplies (IIPS), limited in frequency due to the relatively smooth (exponential) current-voltage characteristic (CVC) of the IIPS bridge diodes through which its capacitor begins to charge, while the spark discharge spectrum is much wider. Further, a signal exceeding the 1st preset level forms a rectangular voltage pulse with a fixed duration, slightly less than a half-period of the supply network, for example, 8 ms, for a network with a voltage frequency of 50 Hz. This duration excludes the influence on the ongoing process of possible “bounce” in the contacts of the switches, if this excess of the signal is caused by a spark discharge in the power contacts of the switch on or off at the moment.

This pulse enters the drive, in the simplest version, it is a single-link integrating RC circuit with constant charge times of 0.1 ... 0.3 s and a discharge of about 0.4 ... 0.6 s. In this case, the voltage level in the drive is compared with the 2nd set level, which, in combination with the charge / discharge time constants of the drive, is selected so that the voltage level in the drive is equal to and exceeds it when a certain number of pulses (4-5) enter the drive with an interval a half-period of the mains supply or a few more pulses with an interval of the network period. If the initiation of the pulse is caused by the switching of the EI, the integrator is then discharged to the initial state, and the switching of 4-5 EI from the protected circuit for a period of 50-100 ms is very unlikely. If a sufficiently intense sparking occurs in the circuit, then the energetics of the process leads to the “occurrence-suppression of the spark” cycle, the main pulse repetition period of which, due to the frequency of the supply voltage, will be equal to the half-period or, in other cases, the period of the mains supply frequency. In this case, the pulses follow continuously and the voltage in the drive exceeds the 2nd set level. This excess initiates the formation of a pulse of a fixed duration arriving at the executive body, disconnecting the protected circuit from the supply network.

Instead of disconnecting the network, the final impulse can also perform (or) an alarm function in a limited class of objects of protection, whose disconnection, for example, requires the approval of the operator.

A block diagram of a device for implementing the method and its placement on the object of protection are presented in figure 1. The device is indicated in FIG. 1 as an ultrasound (spark protection device).

The device operates as follows. The sensor signal passes the RF spectrum forming unit and enters the input of the comparison unit 1, where it is compared with the constant voltage of the first specified level. When the signal exceeds this level at the output of the comparison unit 1, a voltage surge occurs, initiating the start of the unit 1 of the command 1 formation (waiting multivibrator), which generates a pulse of 5 ... 9 ms duration (for an alternating voltage network with a frequency of 50 Hz). This pulse enters the storage unit with a constant charge time of 0.1 ... 0.3 s and a discharge of about 0.4 ... 0.6 s. From the output of the accumulation unit, the signal is fed to the input of the comparison unit 2, where it is compared with the constant voltage of the second predetermined level. When it is exceeded, a voltage jump occurs at the output of the comparison unit 2, which initiates the start of the unit for generating the command 2, which generates a pulse of a fixed duration of the order of 0.5 ... 1 s, which arrives at the executive body, which disconnects the protected circuit from the supply network.

The device can be performed on a serial analog element base, which is based on the well-known dual comparator and dual single-oscillator microcircuits, and therefore has a low weight and dimensions, low power consumption and relative low cost. Devices can be designed and manufactured in one design with well-known circuit breakers, which will expand their functionality and significantly increase their efficiency.

Information sources

1. Patent for invention RU 2159468.

Claims (2)

1. A method for automatically preventing fire from sparking in electrical networks or electrical installations in which the total electric current passing from the electric network to the protected circuit is measured, characterized in that by filtering the measured total current of the protected circuit, low-frequency spectrum components are suppressed to a frequency of several hundred kilohertz, the value of the received signal without accumulation is compared with a predetermined value, and for each excess of this value, a fixed pulse is generated for a long time These are slightly less than the half-period of the supply network, which is integrated in the drive, compared with the threshold achieved when several pulses are received in a drive in succession or with a minimum interval; when the signal level in the drive reaches this threshold, a signal is generated that disconnects the protected electrical circuit from the supply network and ( or) issuing a warning signal, when the pulse arrival rate is less than the specified one, the drive is discharged. 2. Device for automated fire prevention in an electrical network or electrical installation, consisting of a unit for measuring the total electric current passing from the electric network to the protected circuit, an RF spectrum forming unit by filtering the measured total current of the protected circuit and suppressing the low-frequency components of the spectrum to a frequency of several hundreds of kilohertz, the integration unit in the drive, comparisons with the threshold achieved when several pulses in a row arrive at the drive or with minimal m interval, a comparison unit with a given value and for each excess of this value of pulse formation of a fixed duration is slightly less than half the period of the supply network, the command formation block when the signal level in the drive reaches this threshold, the formation of a signal that disconnects the protected electrical circuit from the supply network and generates a warning signal , and an executive body, characterized in that the signal output from the RF spectrum forming unit is connected to the input to an additionally installed unit comparison, the output of which is connected to the input of the additional unit forming the team, and the output from the unit of forming the team is connected to the input of the accumulation unit.