After extensive checkout, it undertook a rendezvous and docking test mission with the unpiloted Shenzhou 8.

Later, during 2012, it conducted docking and crew tests with the crew­carrying Shenzhou 9.

June 2013, Shenzhou 10 joined it in orbit and its crew extended further China's experience in space station type operations.

The chart below is an extract of Tiangong's orbital period from the Twoline Orbital Elements sets issued by Spacetrack. It plots the anomalistic period, or the time it takes to complete one circuit of the orbital ellipse - from perigee to perigee, for example. Some satellite and launch listings give the nodal period which takes account of the fact the ellipse rotates in the plane of the orbit measuring, for example, the time between successive northbound equator crossings. In Tiangong's case, perigee moves forward round the orbit (see the chart below showing argument of perigee) so the nodal period is about 0.1 minute less than the anomalistic period.

Tiangong was injected initially into a slighlty elliptical orbit then it was changed to near-circular.

Apogee and perigee relate to the Twoline Orbital Elements model where the orbit is expressed as an ellipse with the Earth's geographical centre sitting at one focus. The heights are then calculated with respect to the Earth's mean equatorial radius of 6378.145 kilometres. It is an idealised representation and ignores variations in the Earth's shape and the gravitational effect of the Earth's centre of mass being south of the equator. Differing figures published elsewhere may be the same orbit expressed through a different geographical and geometric model.

Perigee is traced in blue and apogee in red.

Within a Twoline Orbital Elements set is a parameter called 'ndot2'. It is a measure of the rate at which an orbit is shrinking. For a low orbit, it is almost exclusively controlled by air drag.

2012 July 10, orbit parameters ceased to follow the usual natural pattern of noticeable fluctuations between element sets. It is best seen in this plot of orbit decay rate where the random ups and downs became a smooth(ish) line. The cause was thruster firings designed to keep the decay rate at a constant level and over-ride short term variations caused by changes in upper atmosphere changes due to Solar activity. Thruster firings ceased September 4.

After Shenzhou 10 departed, Tiangong 1 moved to a higher orbit to prolong its life. Immediately it arrived there 2012 June 26 it began another period of thruster firing to control the rate of orbital decay. On the plot below, the downward-pointing spikes are due to either spurious data from SpaceTrack or the periods when thruster operation ceased for a couple of orbits.

Eccentricity is very sensitive to small changes in orbit parameters. Its tendency is to reduce because air drag causes apogee to decrease at a greater rate than that of perigee unless they are very close to each other in value. Use of thrusters makes eccentricity change suddenly and noticeably, but does not effect apogee and perigee measurements - see the small 'jump' early Oct 10 followed by erratic behaviour. When Tiangong arrived in orbit, the eccentricity was about 0.01. It reduced considerably after the circularisation manoeuvre raised perigee.

The cyclic nature of the changes from the end of 2012 November is due to rotation of the Line of Apsides. The rotation cycle has a period of about 50 days and is driven by variation in the real heights of apogee and perigee above the Geoid. They can differ significantly fom the idealised figures produced by the Twoline Orbital Elements model, and are dependent on the argument of perigee.

The 'spiky' nature of the line during the Shenzhou 9 visit is a side effect of thruster firing to maintain/change attitude, counteract effects of the crew moving around inside, and manoeuvring during the re-docking and undocking phases of the mission. The Shenzhou 8 mission during 2011 did not have the same effect since there was no crew aboard.

Argument of perigee can quite sensitive to the effects of orbital manoeuvres but it depends on the magnitude of the change and whether is occurs at one of the orbital apses (apogee or perigee). When nothing is happening, the argument of perigee increases slowly. An orbit change can to produce a sudden change in argument of perigee. It immediately settles down again to the slow drift.